Advanced Statistical Algorithms Research Articles

Challenges and opportunities beyond structured data in analysis of electronic health records

AbstractElectronic health records (EHR) contain a lot of valuable information about individual patients and the whole population. Besides structured data, unstructured data in EHRs can provide extra, valuable information but the analytics processes are complex, time‐consuming, and often require excessive manual effort. Among unstructured data, clinical text and images are the two most popular and important sources of information. Advanced statistical algorithms in natural language processing, machine learning, deep learning, and radiomics have increasingly been used for analyzing clinical text and images. Although there exist many challenges that have not been fully addressed, which can hinder the use of unstructured data, there are clear opportunities for well‐designed diagnosis and decision support tools that efficiently incorporate both structured and unstructured data for extracting useful information and provide better outcomes. However, access to clinical data is still very restricted due to data sensitivity and ethical issues. Data quality is also an important challenge in which methods for improving data completeness, conformity and plausibility are needed. Further, generalizing and explaining the result of machine learning models are important problems for healthcare, and these are open challenges. A possible solution to improve data quality and accessibility of unstructured data is developing machine learning methods that can generate clinically relevant synthetic data, and accelerating further research on privacy preserving techniques such as deidentification and pseudonymization of clinical text.This article is categorized under: Applications of Computational Statistics > Health and Medical Data/Informatics

Fast and In-Situ Identification of Archaeometallurgical Collections in the Museum of Malaga Using Laser-Induced Breakdown Spectroscopy and a New Mathematical Algorithm

This paper reports the use of an advanced statistical algorithm for the recognition and classification of a set of 30 archaeological metallic objects from the Museum of Malaga. In-situ laser-induced breakdown spectrometry (LIBS) analysis was performed using a portable analyzer. The coordinate-obtaining method provided the statistical weights of each element in the sample. A comparative study between the coordinate-obtaining method and the linear correlation method is also discussed in order to corroborate the applicability of the proposed approach to the field of cultural heritage. The possibility of fast identification based on the simultaneous comparison of all the spectra in the reference LIBS library while allowing the analysis of heterogeneous materials is the main advantage of the method. In addition, statistical analysis (Euclidean distance analysis and binary diagrams) suggested that differentiating between archaeological sites is feasible.

Atomic locations of minor dopants and their roles in the stabilization ofη−Cu6Sn5

Chemical modification using only small amounts of elements such as Zn, In, Sb, or Ni has proven to be an effective means to control the desirable crystal structure of hexagonal η-Cu6Sn5 over a wide thermally operating window, typically found in Pb-free Sn-based soldering or Li-ion battery anode applications. Though appealing, the underlying mechanisms on the role of these dopants remain incomplete and their atomic arrangements within the η-Cu6Sn5 lattices have not yet been experimentally determined. In the current study, we directly reveal the atomic positions of Zn, In, and Sb at the Sn sites of η-Cu6Sn5 via atomic-scale x-ray energy dispersive spectroscopy (XEDS) maps utilizing advanced Cs-corrected scanning transmission electron microscopy. The use of advanced statistical algorithms including Poisson non-local principal component analysis and lattice averaging enables the fine resolution of weak XEDS maps from trace dopant elements. Our first-principles calculations further identify the influence of dopants at these atomic sites on the overall energetics, electronic structures, as well as local bonding environments, leading to the most favorable situations for η-Cu6Sn5 stabilization.

Atomic Locations of Minor Dopants and Their Roles in the Stabilization of <i>η</i>-Cu <sub>6</sub>Sn <sub>5</sub>

Chemical modification using only small amounts of elements such as Zn, In, Sb, or Ni has proven to be an effective means to control the desirable crystal structure of hexagonal η-Cu6Sn5 over a wide thermally operating window, typically found in Pb-free Sn-based soldering or Li-ion battery anode applications. Though appealing, the underlying mechanisms on the role of these dopants remain incomplete and their atomic arrangements within the η-Cu6Sn5 lattices have not yet been experimentally determined. In the current study, we directly reveal the atomic positions of Zn, In, and Sb at the Sn sites of η-Cu6Sn5 via atomic-scale X-ray energy dispersive spectroscopy (XEDS) maps utilizing advanced Cs-corrected scanning transmission electron microscopy (STEM). The use of advanced statistical algorithms including Poisson non-local principal component analysis (Poisson NLPCA) and lattice-averaging enables the fine resolution of weak XEDS maps from trace dopant elements. Our first-principles calculations further identify the influence of dopants at these atomic sites on the overall energetics, electronic structures, as well as local bonding environments, leading to the most favorable situations for η-Cu6Sn5 stabilization.

Novel learning framework (knockoff technique) to evaluate metric ranking algorithms to describe human response to injury

Objective: The objective of this study is to present a novel framework, termed the knockoff technique, to evaluate different metric ranking algorithms to better describe human response to injury.Methods: Many biomechanical metrics are routinely obtained from impact tests using postmortem human surrogates (PMHS) to develop injury risk curves (IRCs). The IRCs form the basis to evaluate human safety in crashworthiness environments. The biomechanical metrics should be chosen based on some measure of their predictive ability. Commonly used algorithms for the choice of ranking the metrics include (a) areas under the receiver operating characteristic curve (AUROC), time-varying AUROC, and other adaptations, and (b) some variants of predictive squared error loss. This article develops a rigorous framework to evaluate the metric selection/ranking algorithms. Actual experimental data are used due to the shortcoming of using simulated data. The knockoff data are meshed into existing experimental data using advanced statistical algorithms. Error rate measures such as false discovery rates (FDRs) and bias are calculated using the knockoff technique. Experimental data are used from previously published whole-body PMHS side impact sled tests. The experiments were conducted at different velocities, padding and rigid load wall conditions, and offsets and with different supplemental restraint systems. The PMHS specimens were subjected to a single lateral impact loading resulting in injury and noninjury outcomes.Results: A total of 25 metrics were used from 42 tests. The AUROC-type algorithms tended to have higher FDRs compared to the squared error loss–type functions (45.3% for the best AUROC-type algorithms versus 31.4% for the best Brier score algorithm). Standard errors for the Brier score algorithm also tended to be lower, indicative of more stable metric choices and robust rankings. The wide variations observed in the performance of the algorithms demonstrated the need for data set–specific evaluation tools such as the knockoff technique developed in this study.Conclusions: In the present data set, the AUROCs and related binary classification algorithms led to inflated FDRs, rendering metric selection/ranking questionable. This is particularly true for data sets with a high proportion of censoring. Squared error loss–type algorithms (such as the Brier score algorithm or its modifications) improved the performance in the metric selection process. The presented new knockoff technique may wholly change how IRCs are developed from impact experiments or simulations. At the very least, the knockoff technique demonstrates the need for evaluations among different metric ranking/selection algorithms, especially when they produce substantially different biomechanical metric choices. Without recommending the AUROC-type or Brier score–type algorithms universally, the authors suggest careful assessments of these algorithms using the proposed framework, so that a robust algorithm may be chosen, with respect to the nature of the experimental data set. Though results are given for sets from a published series of experiments, the knockoff technique is being used by the authors in tests that are applicable to the automotive, aviation, military, and other environments.

A Data-Analytics Tutorial: Building Predictive Models for Oil Production in an Unconventional Shale Reservoir

SummaryConsiderable amounts of data are being generated during the development and operation of unconventional reservoirs. Statistical methods that can provide data-driven insights into production performance are gaining in popularity. Unfortunately, the application of advanced statistical algorithms remains somewhat of a mystery to petroleum engineers and geoscientists. The objective of this paper is to provide some clarity to this issue, focusing on how to build robust predictive models and how to develop decision rules that help identify factors separating good wells from poor performers. The data for this study come from wells completed in the Wolfcamp Shale Formation in the Permian Basin. Data categories used in the study included well location and assorted metrics capturing various aspects of well architecture, well completion, stimulation, and production.Predictive models for the production metric of interest are built using simple regression and other advanced methods such as random forests (RFs), support-vector regression (SVR), gradient-boosting machine (GBM), and multidimensional Kriging. The data-fitting process involves splitting the data into a training set and a test set, building a regression model on the training set and validating it with the test set. Repeated application of a “cross-validation” procedure yields valuable information regarding the robustness of each regression-modeling approach. Furthermore, decision rules that can identify extreme behavior in production wells (i.e., top x% of the wells vs. bottom x%, as ranked by the production metric) are generated using the classification and regression-tree algorithm. The resulting decision tree (DT) provides useful insights regarding what variables (or combinations of variables) can drive production performance into such extreme categories.The main contributions of this paper are to provide guidelines on how to build robust predictive models, and to demonstrate the utility of DTs for identifying factors responsible for good vs. poor wells.

Comprehensive Whole DNA Methylome Analysis by Integrating MeDIP-seq and MRE-seq.

Understanding the role of DNA methylation often requires accurate assessment and comparison of these modifications in a genome-wide fashion. Sequencing-based DNA methylation profiling provides an unprecedented opportunity to map and compare complete DNA CpG methylomes. These include whole genome bisulfite sequencing (WGBS), Reduced-Representation Bisulfite-Sequencing (RRBS), and enrichment-based methods such as MeDIP-seq, MBD-seq, and MRE-seq. An investigator needs a method that is flexible with the quantity of input DNA, provides the appropriate balance among genomic CpG coverage, resolution, quantitative accuracy, and cost, and comes with robust bioinformatics software for analyzing the data. In this chapter, we describe four protocols that combine state-of-the-art experimental strategies with state-of-the-art computational algorithms to achieve this goal. We first introduce two experimental methods that are complementary to each other. MeDIP-seq, or methylation-dependent immunoprecipitation followed by sequencing, uses an anti-methylcytidine antibody to enrich for methylated DNA fragments, and uses massively parallel sequencing to reveal identity of enriched DNA. MRE-seq, or methylation-sensitive restriction enzyme digestion followed by sequencing, relies on a collection of restriction enzymes that recognize CpG containing sequence motifs, but only cut when the CpG is unmethylated. Digested DNA fragments enrich for unmethylated CpGs at their ends, and these CpGs are revealed by massively parallel sequencing. The two computational methods both implement advanced statistical algorithms that integrate MeDIP-seq and MRE-seq data. M&M is a statistical framework to detect differentially methylated regions between two samples. methylCRF is a machine learning framework that predicts CpG methylation levels at single CpG resolution, thus raising the resolution and coverage of MeDIP-seq and MRE-seq to a comparable level of WGBS, but only incurring a cost of less than 5% of WGBS. Together these methods form an effective, robust, and affordable platform for the investigation of genome-wide DNA methylation.

Qualitative Assessments via Infrared Vision of Sub-surface Defects Present Beneath Decorative Surface Coatings

In this work, the potentialities of the infrared vision to explore sub-superficial defects in polychromatic statues were investigated. In particular, it was possible to understand how the reflector effect of the exterior golden layers could be minimized, applying advanced statistical algorithms to thermal images. Since this noble metal is present as external coating in both artworks, an in-depth discussion concerning its physicochemical properties is also added. In this context, the principal component thermography technique and, the more recent, partial least squares thermography technique were used on three different datasets recorded, providing long thermal stimuli. The main images were compared both to phasegrams and to the thermographic signal reconstruction results in order to have a clear outline of the situation to be debated. The effects of view factors on the radiation transfer linked to the specular reflections from the surface did not falsely highlight certain features inadvertently. Indeed, the raw thermograms were analyzed one by one. Reflectograms were used to pinpoint emissivity variations due to, e.g., possible repainting. The paper concludes that, as it is possible to understand from a physical point of view, the near-infrared reflectography technique is able to examine the state of conservation of the upper layers in cultural heritage objects, while the infrared thermography technique explores them more in-depth. The thesis statement is based on the thermal and nonthermal parts of the infrared region, therefore, indicating what can be detected by heating the surface and what can be visualized by illuminating the surface, bearing in mind the nature of the external coating.

IQuant: an automated pipeline for quantitative proteomics based upon isobaric tags.

Quantitative proteomics technology based on isobaric tags is playing an important role in proteomic investigation. In this paper, we present an automated software, named IQuant, which integrates a postprocessing tool of protein identification and advanced statistical algorithms to process the MS/MS signals generated from the peptides labeled by isobaric tags and aims at proteomics quantification. The software of IQuant, which is freely downloaded at http://sourceforge.net/projects/iquant/, can run from a graphical user interface and a command-line interface, and can work on both Windows and Linux systems.

Indirect Performance Sensing for On-Chip Self-Healing of Analog and RF Circuits

The advent of the nanoscale integrated circuit (IC) technology makes high performance analog and RF circuits increasingly susceptible to large-scale process variations. On-chip self-healing has been proposed as a promising remedy to address the variability issue. The key idea of on-chip self-healing is to adaptively adjust a set of on-chip tuning knobs (e.g., bias voltage) in order to satisfy all performance specifications. One major challenge with on-chip self-healing is to efficiently implement on-chip sensors to accurately measure various analog and RF performance metrics. In this paper, we propose a novel indirect performance sensing technique to facilitate inexpensive-yet-accurate on-chip performance measurement. Towards this goal, several advanced statistical algorithms (i.e., sparse regression and Bayesian inference) are adopted from the statistics community. A 25 GHz differential Colpitts voltage-controlled oscillator (VCO) designed in a 32 nm CMOS SOI process is used to validate the proposed indirect performance sensing and self-healing methodology. Our silicon measurement results demonstrate that the parametric yield of the VCO is significantly improved for a wafer after the proposed self-healing is applied.

160-P: neXtype: A SAFE CAR ON THE NGS ROLLER COASTER

Aim NeXtype is a software aiming at high-accuracy HLA typing based on short amplicon Next Generation Sequencing (NGS) data from Illumina MiSeq systems. Methods NeXtype employs advanced statistical algorithms that use high number of reads and associated Q-values delivered by NGS. In our NGS setup, MiSeq provides raw sequence data of paired-end reads. For all samples, 6 loci are considered on exons 2 and 3. The neXtype workflow is organized in several consecutive steps. Up-front, pre-run data is assembled, including a tree-structured HLA library and primer information. In the first step of an individual typing run, the primer sequence is identified for each read by a multi-level process that involves direct string comparison and also adapts to sequencing errors. Next, for each pair of reads the most likely matches in the HLA library are determined using Q-values and the predefined tree structure. The third step evaluates the linkage between all matches of each locus and exon and orders them by significance. The fourth step classifies the called alleles as results or discards them as crossover products, statistical noise, co-amplifications etc. For each locus, allele pairs are obtained by combining the respective exon information. Although neXtype reliably rates its results, a user can override the result classification if deemed necessary and declare results as artifacts for future reference. Results NGS results by neXtype have successfully been cross-validated with Sanger sequencing results of 6840 loci. In average, 98.7% of all typings have high resolution. 94.9% do not require user interaction, 2.7% require user editing for improvement of the result, and in 2.4 % data quality was insufficient to allow HLA typing. This high degree of automation reduces analysis time to less than 2 h per 2000 loci. Conclusions NeXtype matches the needs of high throughput HLA labs providing accurate typing with minimal user intervention. Currently, 9,000 loci are typed per day.