Multiple Imputation Algorithm Research Articles

#1809 Twice against thrice-weekly hemodialysis: the TATH trial

Abstract Background and Aims Frequency of hemodialysis is still a debatable topic. Incremental hemodialysis has gained popularity and requires a smooth increase in dialysis frequency while maintaining a sufficient residual diuresis. In developing countries and some developed countries, a twice-weekly schedule independent of residual kidney function is still performed, sometimes because of lack of resources and some other times because of patients’ resistance to undergo three sessions per week. This study aimed to compare the survival of patients on twice versus thrice-weekly hemodialysis. Method This is a prospective, multicenter, non-randomized trial, that included incident patients on chronic hemodialysis between January 2018 and August 2021. Patients terminally ill were excluded. Patients were divided into two arms: twice-weekly and thrice-weekly hemodialysis. CASTOR eCRF (electronic case report form) was used for data capture. Demographics, comorbidities, laboratory tests, medications, residual diuresis and dialysis prescription were collected from the patients' medical records at dialysis initiation. Patients were followed at 1 month, 3 months, 6 months, one year and two years for their mortality, hospitalisations for pulmonary edema or hyperkalemia, uncontrolled severe hypertension and erythropoietin (EPO) dose requirement. This trial was terminated before reaching the required sample size of 800 because of the pandemic and economic crisis in the country. Missing baseline factors and covariates were imputed using multiple imputation algorithms, then entered in a logistic regression model to estimate propensity scores. A survival analysis with a cox-regression model including propensity score as a covariate assessed the difference between twice-weekly and thrice-weekly dialysis. This trial was registered on Clinicaltrials.gov ID NCT03415776. Results A total of 203 consecutive patients of 10 dialysis units were included, 133 in the thrice-weekly arm and 70 in the twice-weekly. Mean age was 67 ± 15 years, 54.2% were males, 55.5% had diabetes, 53.7% had coronary artery disease. Median eGFR at dialysis initiation was 6 (4, 8) mL/min/1.73 m2. Baseline median 24-hour diuresis was 500 (500, 1000) mL with higher diuresis in patients on twice-weekly (P = 0.051). Baseline mean hemoglobin was 9.6 ± 1.6 g/dL. 41.7% were initiated with AV fistula with more patients in the thrice-weekly initiated with temporary catheter. Median number of dialysis hours per week was 12 (10, 12) and 8 (8, 9) in thrice- and twice-weekly respectively. High-flux membranes with ultrapure dialysate were used in 55% of thrice-weekly and 74.3% of twice-weekly. Mean blood flow was 329 ± 76 mL/min. In univariate analysis, residual diuresis was significantly higher in twice-weekly at 1 month, 3 months (P = .002), 6 months (it dropped at 6 months to a median of 100 (0, 500) in thrice-weekly versus 225 (175, 875) in twice-weekly) and at one year (P = .015). Inter-dialytic weight gain and blood pressure levels were not statistically different at 1, 3, 6 months and one year. 22.5% of patients on thrice-weekly did not need antihypertensive medications at 6 months versus 12.1% in the twice-weekly. Serum potassium was significantly higher in twice-weekly at 3 months and at one year. Cumulative dose of EPO administered during 2 years was significantly higher in twice-weekly with a median of 720000 UI versus 500000 UI in thrice-weekly (P = .001) without a difference in hemoglobin but lower ferritin in twice-weekly (P = .005). There was no difference in hospitalisations' rates between the two arms. A cox regression model including propensity scores as a covariate did not show any survival difference between the two arms (Fig. 1). Conclusion Patients on twice-weekly hemodialysis had a similar survival and hospitalisations' rates at two years as patients on thrice-weekly hemodialysis. However, a twice-weekly schedule was accompanied with higher EPO dose requirements, higher serum potassium levels, more antihypertensive therapy despite more preserved residual diuresis.

Advanced machine learning for missing petrophysical property imputation applied to improve the characterization of carbonate reservoirs

Missing data is a common cause of uncertainty in reservoir characterization, especially in core analysis of porosity and permeability. This is due to the high cost and time required to gather core samples while drilling through the entire reservoir thickness. Therefore, the ability to collect core samples and directly assess petrophysical parameters is restricted by the small proportion of the reservoir thickness that can be measured by the retrieved cores. As a result, the absence of data has the potential to lead to inconsistent and imprecise classification of reservoirs and geomodeling, thereby increasing reservoir uncertainty. Therefore, imputation algorithms are needed to estimate petrophysical properties for incomplete intervals. This study comprehensively compares seven imputation techniques for predicting missing horizontal and vertical core permeability and core porosity data in a well drilled through a carbonate reservoir in a southern Iraqi gas field. The performance of each method was assessed using relative bias (RB) and reasonable descriptive statistical distributions (histogram, log view, and box plots of data ratios to compare the imputed and original data), and robustness to outliers (RtO). The results revealed that the random imputation of missing data (RIMD) technique is the most effective algorithm in addressing the 30.5% missing data (vertical permeability), resulting in the attainment of the RB value closest to zero (0.04076191) compared to other algorithms. Principal component analysis (PCA) and random forest (RF) yielded the most favorable results in terms of handling missing values and outliers for the 15% missing ratio dataset (core porosity and horizontal permeability). The advantages of these approaches are related to their capacity to independently and accurately impute the missing data without requiring extra information, such as well logging records. Additionally, the multiple imputation and machine learning algorithms fill missing values with adequate numbers based on dataset variable distributions and correlations. Multiple plausible values quantify the uncertainty in estimating missing data, avoiding false precision (as with single imputation). Also, the novel workflow involving multiple imputation methods accurately take into account the correlations and variances between the variables. The novel workflow developed for missing data imputation has the potential to be applied to other clastic and carbonate reservoirs in oil and gas fields. The entire data imputation workflow was conducted, validated, visualized by codes in R, a robust open-source statistical computing software.

Abstract PO1-17-07: Invasive disease-free survival as a surrogate for overall survival in patients with hormone receptor−positive/human epidermal growth factor receptor 2−negative early breast cancer: a real-world analysis

Abstract Background: Although the goal of treatment is to improve overall survival (OS), in the adjuvant setting, disease-free survival (DFS) is an important endpoint because DFS risk reduction may reduce mortality risk. After introduction of the STEEP criteria, invasive DFS (iDFS) was adopted as an objective, well accepted primary endpoint in early breast cancer (EBC) trials. This analysis evaluated iDFS as a surrogate for OS in the HR+/HER2− EBC adjuvant treatment setting using real-world data. Methods: This retrospective analysis of the ConcertAI Patient360 database contains deidentified electronic medical records of patients (pts) treated at US academic and community oncology clinics from Jan 1, 1995 to Apr 30, 2021. The cohort included pts with AJCC (8th ed.) stage II-III (if IIIB or IIIC, confirmation was required on residual tumor status; pathological and/or clinical staging [pathological stage was prioritized if both existed in the record]) HR+/HER2− EBC who had surgery and initiated adjuvant endocrine therapy (ET); ovarian function suppression was permitted. iDFS (disease recurrence, metastasis, second primary tumor, or death) and OS (death due to any cause) were defined as time between ET start and first qualifying event. Pts not experiencing an event were censored at data cutoff or maximum follow-up (whichever was first) for iDFS and maximum follow-up for OS. Correlation analyses between individual iDFS and OS times were performed using Pearson and Spearman correlations and an iterative multiple imputation (IMI) algorithm. Subgroup analyses were conducted by first ET, menopausal status (imputed as 50 years if missing), stage, nodal status, prior (neo)adjuvant chemotherapy, and prior radiation. Results: In total, 3133 pts were analyzed. Mean (SD) age (index or ET initiation date) was 58.4 (12.4) years; 98.8% of pts were female, 29.9% were premenopausal, 80.9% had stage II disease, 57.1% had nodal involvement, and 41.6% and 48.8% had received (neo)adjuvant chemotherapy and radiation therapy. Overall, 1854 and 653 pts received nonsteroidal aromatase inhibitors and tamoxifen only. The median follow-up time was 55.1 months for iDFS and 68.2 months for OS. The iDFS (1103 events [35.2%]) rates were 88.9% and 73.9% at 2 and 5 years. The OS (554 events [17.7%]) rates were 97.4% and 90.5% at 2 and 5 years. There was a significant and high correlation ( >0.85 based on IQWiG 2011 guidelines) between iDFS and OS in the overall cohort (Pearson 0.91, P< .001; Spearman 0.88, P< .001) with 82% of OS variation explained by iDFS (least square R2=0.82; Table). Results were confirmed by IMI rho (0.83, P< .001), which is interpreted as very strong ( >0.8). Results were consistent among all reported subgroups; Pearson and Spearman correlations exceeded 0.85 and IMI exceeded 0.8, except for pts with stage III disease where correlation remained high (Pearson 0.88, P< .001) or medium/strong (Spearman 0.84, P< .001; IMI 0.79, P< .001) and where most censoring was observed (38.2% of pts). Conclusions: This retrospective cohort analysis demonstrates a very strong pt-level surrogacy between iDFS and OS among pts with HR+/HER2− EBC treated in a real-world, US-based, academic and community setting. The findings complement the trial-level surrogacy from prior analyses using data from randomized controlled trials and support iDFS as a surrogate endpoint for OS in HR+/HER2− EBC trials. Citation Format: Stephanie Graff, Sara Tolaney, Lowell Hart, Pedram Razavi, Wolfgang Janni, Lee Schwartzberg, Andrii Danyliv, Juan Carlos Mora Payan, Ilia Ferrusi, Rishi Rajat Adhikary, Joyce O'Shaughnessy. Invasive disease-free survival as a surrogate for overall survival in patients with hormone receptor−positive/human epidermal growth factor receptor 2−negative early breast cancer: a real-world analysis [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-17-07.

Variable selection for high-dimensional incomplete data

Regression analysis is often affected by high dimensionality, severe multicollinearity, and a large proportion of missing data. These problems may mask important relationships and even lead to biased conclusions. This paper proposes a novel computationally efficient method that integrates data imputation and variable selection to address these issues. More specifically, the proposed method incorporates a new multiple imputation algorithm based on matrix completion (Multiple Accelerated Inexact Soft-Impute), a more stable and accurate new randomized lasso method (Hybrid Random Lasso), and a consistent method to integrate a variable selection method with multiple imputation. Compared to existing methodologies, the proposed approach offers greater accuracy and consistency through mechanisms that enhances robustness against different missing data patterns and sampling variations. The method is applied to analyze the Asian American minority subgroup in the 2017 National Youth Risk Behavior Survey, where key risk factors related to the intention for suicide among Asian Americans are studied. Through simulations and real data analyses on various regression and classification settings, the proposed method demonstrates enhanced accuracy, consistency, and efficiency in both variable selection and prediction.

Handling intercurrent events and missing data in non-inferiority trials using the estimand framework: A tuberculosis case study.

The ICH E9 addendum outlining the estimand framework for clinical trials was published in 2019 but provides limited guidance around how to handle intercurrent events for non-inferiority studies. Once an estimand is defined, it is also unclear how to deal with missing values using principled analyses for non-inferiority studies. Using a tuberculosis clinical trial as a case study, we propose a primary estimand, and an additional estimand suitable for non-inferiority studies. For estimation, multiple imputation methods that align with the estimands for both primary and sensitivity analysis are proposed. We demonstrate estimation methods using the twofold fully conditional specification multiple imputation algorithm and then extend and use reference-based multiple imputation for a binary outcome to target the relevant estimands, proposing sensitivity analyses under each. We compare the results from using these multiple imputation methods with those from the original study. Consistent with the ICH E9 addendum, estimands can be constructed for a non-inferiority trial which improves on the per-protocol/intention-to-treat-type analysis population previously advocated, involving respectively a hypothetical or treatment policy strategy to handle relevant intercurrent events. Results from using the 'twofold' multiple imputation approach to estimate the primary hypothetical estimand, and using reference-based methods for an additional treatment policy estimand, including sensitivity analyses to handle the missing data, were consistent with the original study's reported per-protocol and intention-to-treat analysis in failing to demonstrate non-inferiority. Using carefully constructed estimands and appropriate primary and sensitivity estimators, using all the information available, results in a more principled and statistically rigorous approach to analysis. Doing so provides an accurate interpretation of the estimand.

Multiple imputation of more than one environmental exposure with nondifferential measurement error.

Measurement error is common in environmental epidemiologic studies, but methods for correcting measurement error in regression models with multiple environmental exposures as covariates have not been well investigated. We consider a multiple imputation approach, combining external or internal calibration samples that contain information on both true and error-prone exposures with the main study data of multiple exposures measured with error. We propose a constrained chained equations multiple imputation (CEMI) algorithm that places constraints on the imputation model parameters in the chained equations imputation based on the assumptions of strong nondifferential measurement error. We also extend the constrained CEMI method to accommodate nondetects in the error-prone exposures in the main study data. We estimate the variance of the regression coefficients using the bootstrap with two imputations of each bootstrapped sample. The constrained CEMI method is shown by simulations to outperform existing methods, namely the method that ignores measurement error, classical calibration, and regression prediction, yielding estimated regression coefficients with smaller bias and confidence intervals with coverage close to the nominal level. We apply the proposed method to the Neighborhood Asthma and Allergy Study to investigate the associations between the concentrations of multiple indoor allergens and the fractional exhaled nitric oxide level among asthmatic children in New York City. The constrained CEMI method can be implemented by imposing constraints on the imputation matrix using the mice and bootImpute packages in R.

MIVAE: Multiple Imputation based on Variational Auto-Encoder

Nowadays, the issue of MV imputation has become one of the research hotspots in the field of data quality, since the missing values (MVs) are prevalent in real-world datasets and bring challenges to advanced data analytics algorithms. To impute the MVs, most existing approaches directly derive one estimation for each MV, which is categorized as the single imputation (SI). However, the SI ignores the uncertainty of the MVs, and thereby usually derive unsatisfactory imputation results compared to the Multiple imputation (MI). To extract the uncertainty of the MVs, the MI algorithms derive multiple candidate estimations for each MV. Nevertheless, existing MI approaches are few due to the complicated data-handling process. Accordingly, in this paper, by exploring the Variational Auto-Encoder (VAE) model, we propose a new MI approach, namely MIVAE (Multiple Imputation based on Variational Auto-Encoder) to impute MVs for the tabular data. In MIVAE, we first add a corrupted input layer (where the synthetic MVs are introduced) adjacent to the original input layer to make the model capable of MV issue. Then, we obtain multiple rather than single candidate estimations for each data sample from the posterior distribution of the latent variables learned by our designed model. In such way, the multiple imputation is effectively implemented where the uncertainty of the MVs are extracted perfectly. Next, to obtain satisfactory imputation results, we add a data analysis layer at the end of the network to integrate multiple candidate estimations intelligently. Finally, the experimental results over four real-world datasets demonstrate that MIVAE achieves significantly higher imputation accuracy compared to existing solutions, and MIVAE are capable of handling both numerical and categorized tabular data. For example, the imputation accuracy based on MIVAE improves up to about 40% and 30% compared with PMM and MIWAE (which are the state-of-the-art MI approach) over the CropMapping dataset, respectively. Moreover, we train a MIVAE model over three datasets containing MVs, respectively. By leveraging the trained MIVAE, the classification performance over the imputed data is similar to that over the complete data.

A method for comparing multiple imputation techniques: A case study on the U.S. national COVID cohort collaborative

Healthcare datasets obtained from Electronic Health Records have proven to be extremely useful for assessing associations between patients’ predictors and outcomes of interest. However, these datasets often suffer from missing values in a high proportion of cases, whose removal may introduce severe bias. Several multiple imputation algorithms have been proposed to attempt to recover the missing information under an assumed missingness mechanism. Each algorithm presents strengths and weaknesses, and there is currently no consensus on which multiple imputation algorithm works best in a given scenario. Furthermore, the selection of each algorithm’s parameters and data-related modeling choices are also both crucial and challenging.In this paper we propose a novel framework to numerically evaluate strategies for handling missing data in the context of statistical analysis, with a particular focus on multiple imputation techniques. We demonstrate the feasibility of our approach on a large cohort of type-2 diabetes patients provided by the National COVID Cohort Collaborative (N3C) Enclave, where we explored the influence of various patient characteristics on outcomes related to COVID-19. Our analysis included classic multiple imputation techniques as well as simple complete-case Inverse Probability Weighted models. Extensive experiments show that our approach can effectively highlight the most promising and performant missing-data handling strategy for our case study. Moreover, our methodology allowed a better understanding of the behavior of the different models and of how it changed as we modified their parameters.Our method is general and can be applied to different research fields and on datasets containing heterogeneous types.

A Noise-Aware Multiple Imputation Algorithm for Missing Data

Missing data is a common and inevitable phenomenon. In practical applications, the datasets usually contain noises for various reasons. Most of the existing missing data imputing algorithms are affected by noises which reduce the accuracy of the imputation. This paper proposes a noise-aware missing data multiple imputation algorithm NPMI in static data. Different multiple imputation models are proposed according to the missing mechanism of data. Secondly, the method to determine the imputation order of multivariablesmissing is given. A random sampling consistency algorithm is proposed to estimate the initial values of the parameters of the multiple imputation model to reduce the influence of noise data and improve the algorithm’s robustness. Experiments on two real datasets and two synthetic datasets verify the accuracy and efficiency of the proposed NPMI algorithm, and the results are analyzed.

An evaluation of methods to handle missing data in the context of latent variable interaction analysis: multiple imputation, maximum likelihood, and random forest algorithm

An evaluation of methods to handle missing data in the context of latent variable interaction analysis: multiple imputation, maximum likelihood, and random forest algorithm

Globalization spatial spillover effects and carbon emissions: What is the role of economic complexity?

This paper employs the spatial panel model to explore the interaction between globalization and economic complexity on CO2 emissions. To capture different potential mechanisms of diffusions, this paper builds hybrid W matrices based on trade distance and geographical distances. Further, to reduce small sample bias and capture “true interactions” across countries, the multiple imputation algorithm is used to address missingness pattern. This approach can efficiently enhance data quality and inferences validity. Furthermore, to distinguish the spatial spillover accurately, cross-sectional averages are used to solve the issue caused by the omission of common factors or shocks that affect differently the different spatial units. Results show that there is a positive spatial spillover of CO2 emissions from neighboring countries to the local country. Furthermore, the results show that globalization on its own has no significant effect on CO2 emissions in the local country but decreases CO2 emissions in neighboring countries. More importantly, when taking economic complexity into account, high level of economic complexity may decrease the negative indirect impact of globalization on carbon emissions. The more complex countries are to the economic structure, the less is the negative impact of globalization on carbon emissions in spatially related countries. These findings highlight that the role of economic complexity and spatial spillovers effects are imperative to be considered. Based on this study, several policy recommendations are provided.

Infinite hidden Markov models for multiple multivariate time series with missing data.

Exposure to air pollution is associated with increased morbidity and mortality. Recent technological advancements permit the collection of time-resolved personal exposure data. Such data are often incomplete with missing observations and exposures below the limit of detection, which limit their use in health effects studies. In this paper, we develop an infinite hidden Markov model for multiple asynchronous multivariate time series with missing data. Our model is designed to include covariates that can inform transitions among hidden states. We implement beam sampling, a combination of slice sampling and dynamic programming, to sample the hidden states, and a Bayesian multiple imputation algorithm to impute missing data. In simulation studies, our model excels in estimating hidden states and state-specific means and imputing observations that are missing at random or below the limit of detection. We validate our imputation approach on data from the Fort Collins Commuter Study. We show that the estimated hidden states improve imputations for data that are missing at random compared to existing approaches. In a case study of the Fort Collins Commuter Study, we describe the inferential gains obtained from our model including improved imputation of missing data and the ability to identify shared patterns in activity and exposure among repeated sampling days for individuals and among distinctindividuals.

Equivalent imputation methodology for handling missing data in compositional geochemical databases of geothermal fluids

Chemical, isotopic and thermal signatures, space-temporal concentration patterns, and geochemical conceptual models are used to gather insights of the internal structure of geothermal reservoirs. These investigations rely on historical and latest data dealing with the composition of the geothermal fluids discharged from either producer wells or surface manifestations. The creation of a compositional database is based on information compiled from the sample collection and handling, and the quantification results of chemical analyses. Sometimes, this integrated task may result incomplete due to some technical issues that might generate information gaps or missing data which are struggle to control in the field and lab works. To solve the low-quality data, the missing concentration values or the significant information gaps of geochemical databases, the use of machine learning techniques has been explored in the present study. Equivalent imputation models were proposed, for the first time, to rebuild a world solute-based geochemical database which exhibited some gaps and missing data in the records of the chemical composition of fluids from geothermal production wells. The application of these models enabled that missing records to be successfully recovered by using single and multiple imputation algorithms. The imputation models were efficiently evaluated by applying statistical equivalence metrics carried out between imputed and observed (measured) data. To avoid an arbitrary use of these models, and to keep the original geochemical signatures of the fluid composition patterns, the imputed results were assessed by conventional geochemistry tools (e.g., charge unbalances, fluid classification, and some water–rock interaction footprints). Results show that single imputation algorithms that model locally provide equivalent imputation values with acceptable confidence, not only in terms of statistical metrics but also improving charge unbalances in chemical analyses of anomalous samples without altering the original water–rock interaction signatures of the geothermal systems under evaluation. In contrast, multiple imputation algorithms only tend to obtain better charge unbalances by altering the original water–rock interaction signatures. The novel use of the imputation techniques for filling geochemical databases and recovering missing data will allow a better interpretation of hydrogeochemistry modeling in future geothermal exploration and exploitation studies.

Intelligent Data Processing Methods for the Atypical Values Correction of Stock Quotes

Purpose of the study. The purpose of the study is to carry out a comparative analysis of various methods for correcting atypical values of statistical data on the stock market and to develop recommendations for their use.Materials and methods. The article analyzes Russian and foreign bibliography on the research problem. Consideration of machine learning methods for detecting and correcting outliers in time series is proposed. The mathematical basis of machine learning methods is the Z-score method, the isolation forest method, support vector method for outlier detection, and winsorization and multiple imputation methods for outlier correction. To create the models, the Jupyter Notebook software tool, which supports the Python programming language, was used. To implement machine-learning methods, data from stock quotes of the Moscow Exchange are used.Results. The results of machine learning algorithms are demonstrated for sets of real statistical data representing the closing prices of shares of three Russian companies “Sberbank”, “Aeroflot”, “Gazprom” in the period from 01.12.2019 to 30.11.2020, obtained from the website of the Investment Company “FINAM”. A comparative analysis of methods for detecting and correcting outliers by standard deviation has been carried out. The Z-score statistical method allows you to accurately determine the distance from the suspicious observation to the distribution center, which is an advantage. The disadvantage of this method is the influence of outliers on the mean and standard deviation, which can contribute to the masking of outliers or their incorrect detection. The isolation forest method recognizes outliers of various types, and when implementing the method, there are no parameters that require selection; but the disadvantage is the slower detection rate of outliers compared to other methods. The support vector machine is a very fast method and is reduced to solving a quadratic programming problem, which always has a unique solution. The winsorization method for correcting outliers reduces the effect of outliers on the mean and variance, which is an advantage, but may introduce bias due to the selection of thresholds to separate observations in the sample. The multiple imputation method creates for each missing value not one, but many imputations, which avoids a systematic error, but at the expense of high computational costs. For the initial data used in the work, the best result was shown by the implementation of the multiple imputation algorithm based on the detected outliers by the support vector method.Conclusion. There is no universal method for detecting and/or eliminating outliers in data analysis theory. In general, the determination of outliers is subjective, and the decision is made individually for each specific dataset, considering its characteristics or existing experience in this area. The practical implementation of the methods for detecting and eliminating outliers used in this work can be a tool for calculating more accurate indicators in any area, for example, to improve forecasting the stock price. As part of further work, it is possible to consider the optimization of the parameters used in the methods of detecting and correcting outliers to study their effect on the results of the models.

A novel model to optimize multiple imputation algorithm for missing data using evolution methods

The concept of missing data is considered significant when applying statistical methods to a dataset and the quality of the data analysis results is based on the correct data completeness. As a result, improving missing data filling processes is vital in order to give more reliable data throughout the phase of analysis. Here, we present a novel method for optimizing multiple regression imputation processes and obtaining the best fitness values for missing data from patients by combining multiple imputations with a genetic algorithm. To train and assess our proposed method, we employed 583 patient records from a publicly available database, divided into 416 records of liver patients and 167 records of the non-liver patients. The proposed approach offers the largest improvement for missing data findings, according to the results. Instead of employing the normal equation in multiple imputations, which yielded 92.72 as the utmost fitness value with Mean Absolute Error (MAE) 0.5877 from 1.1840 after our second optimization, we were able to achieve a fitness value of 233. The proposed approach might be tested using a large database and used in Hepatocellular carcinoma (HCC) labs to help clinicians make accurate diagnoses.

Impact of different cover letter content and incentives on non-response bias in a sample of Veterans applying for Department of Veterans Affairs disability benefits: a randomized, 3X2X2 factorial trial

BackgroundNon-random non-response bias in surveys requires time-consuming, complicated, post-survey analyses. Our goal was to see if modifying cover letter information would prevent non-random non-response bias altogether. Our secondary goal tested whether larger incentives would reduce non-response bias.MethodsA mailed, survey of 480 male and 480 female, nationally representative, Operations Enduring Freedom, Iraqi Freedom, or New Dawn (OEF/OIF/OND) Veterans applying for Department of Veterans Affairs (VA) disability benefits for posttraumatic stress disorder (PTSD). Cover letters conveyed different information about the survey’s topics (combat, unwanted sexual attention, or lifetime and military experiences), how Veterans’ names had been selected (list of OEF/OIF/OND Veterans or list of Veterans applying for disability benefits), and what incentive Veterans would receive ($20 or $40). The main outcome, non-response bias, measured differences between survey respondents’ and sampling frame’s characteristics on 8 administrative variables, including Veterans’ receipt of VA disability benefits and exposure to combat or military sexual trauma. Analysis was intention to treat. We used ANOVA for factorial block-design, logistic, mixed-models to assess bias and multiple imputation and expectation-maximization algorithms to assess potential missing mechanisms (missing completely at random, missing at random, or not random) of two self-reported variables: combat and military sexual assault.ResultsRegardless of intervention, men with any VA disability benefits, women with PTSD disability benefits, and women with combat exposure were over-represented among respondents. Interventions explained 0.0 to 31.2% of men’s variance and 0.6 to 30.5% of women’s variance in combat non-response bias and 10.2 to 43.0% of men’s variance and 0.4 to 31.9% of women’s variance in military sexual trauma non-response bias. Non-random assumptions showed that men’s self-reported combat exposure was overestimated by 19.0 to 28.8 percentage points and their self-reported military sexual assault exposure was underestimated by 14.2 to 28.4 percentage points compared to random missingness assumptions. Women’s self-reported combat exposure was overestimated by 8.6 to 10.6 percentage points and military sexual assault exposure, by 1.2 to 6.9 percentage points.ConclusionsOur interventions reduced bias in some characteristics, leaving others unaffected or exacerbated. Regardless of topic, researchers are urged to present estimates that include all three assumptions of missingness.

A novel optimization algorithm for the missing data in HCC based on multiple imputation and genetic algorithm

A novel optimization algorithm for the missing data in HCC based on multiple imputation and genetic algorithm

Comparison of multiple imputation algorithms and verification using whole-genome sequencing in the CMUH genetic biobank.

A genome-wide association study (GWAS) can be conducted to systematically analyze the contributions of genetic factors to a wide variety of complex diseases. Nevertheless, existing GWASs have provided highly ethnic specific data. Accordingly, to provide data specific to Taiwan, we established a large-scale genetic database in a single medical institution at the China Medical University Hospital. With current technological limitations, microarray analysis can detect only a limited number of single-nucleotide polymorphisms (SNPs) with a minor allele frequency of >1%. Nevertheless, imputation represents a useful alternative means of expanding data. In this study, we compared four imputation algorithms in terms of various metrics. We observed that among the compared algorithms, Beagle5.2 achieved the fastest calculation speed, smallest storage space, highest specificity, and highest number of high-quality variants. We obtained 15,277,414 high-quality variants in 175,871 people by using Beagle5.2. In our internal verification process, Beagle5.2 exhibited an accuracy rate of up to 98.75%. We also conducted external verification. Our imputed variants had a 79.91% mapping rate and 90.41% accuracy. These results will be combined with clinical data in future research. We have made the results available for researchers to use in formulating imputation algorithms, in addition to establishing a complete SNP database for GWAS and PRS researchers. We believe that these data can help improve overall medical capabilities, particularly precision medicine, in Taiwan.

Comparison of Missing Data Infilling Mechanisms for Recovering a Real-World Single Station Streamflow Observation.

Reconstructing missing streamflow data can be challenging when additional data are not available, and missing data imputation of real-world datasets to investigate how to ascertain the accuracy of imputation algorithms for these datasets are lacking. This study investigated the necessary complexity of missing data reconstruction schemes to obtain the relevant results for a real-world single station streamflow observation to facilitate its further use. This investigation was implemented by applying different missing data mechanisms spanning from univariate algorithms to multiple imputation methods accustomed to multivariate data taking time as an explicit variable. The performance accuracy of these schemes was assessed using the total error measurement (TEM) and a recommended localized error measurement (LEM) in this study. The results show that univariate missing value algorithms, which are specially developed to handle univariate time series, provide satisfactory results, but the ones which provide the best results are usually time and computationally intensive. Also, multiple imputation algorithms which consider the surrounding observed values and/or which can understand the characteristics of the data provide similar results to the univariate missing data algorithms and, in some cases, perform better without the added time and computational downsides when time is taken as an explicit variable. Furthermore, the LEM would be especially useful when the missing data are in specific portions of the dataset or where very large gaps of ‘missingness’ occur. Finally, proper handling of missing values of real-world hydroclimatic datasets depends on imputing and extensive study of the particular dataset to be imputed.

Importance sampling imputation algorithms in quantile regression with their application in CGSS data

As a popular sampling tool used for Monte Carlo computing, importance sampling (IS) has been used in a wide variety of application areas recently. In large-scale survey data such as Chinese General Social Survey (CGSS), missing data, a high level of skewness and heteroscedastic variances commonly occur. Most of the existing literatures on survey data analysis focused on modeling the conditional mean without deep investigation about missing data. In this paper, we study an IS imputation algorithm and its modified algorithm with inverse probability weighting arrangement in quantile regression with missing covariates. We make full use of the observed data compared with the existing complete cases analysis and inverse probability weighting method, and also improve the computational efficiency of multiple imputation and EM-based algorithms. The quantile regression framework allows us to obtain an overall picture of covariates’ effects, highlights the changing relationships according to the explored quantile of interest, and solves the problems of skewness and heterogeneity. Through simulation studies, we investigate the performances of both IS and ISW with other existing algorithms. Finally, we apply our algorithms to part of annual income data from CGSS in 2010. We build three kinds of quantile regression models based on all the subjects, urban areas subjects and rural areas subjects, respectively.