NT-proBNP for Predicting All-Cause Death and Heart Transplant in Children and Adults with Heart Failure.

Plasma N-terminal prohormone B-type natriuretic peptide (NT-proBNP) concentration is a heart failure (HF) biomarker in adults and children. Its prognostic value for HF-related events has been established only in adults. Therefore, we aimed to test the hypothesis that plasma NT-proBNP concentrations predicted the risk of heart transplantation or death in children with HF. We studied the medical records of 109 children with HF enrolled in the IBM Watson Explorys database and from 150 children enrolled in the Pediatric Cardiomyopathy Registry (PCMR). Nonlinear regression was used to assess the relationship between plasma NT-proBNP concentrations and the risk of events in the two cohorts. All children in the PCMR cohort had dilated cardiomyopathy. The Explorys cohort also included children with congenital cardiovascular malformations. Median plasma NT-proBNP concentrations were 1250 pg/mL and 184 pg/mL in the Explorys and PCMR cohorts, respectively. The percentage of deaths/heart transplantations was 7%/22%, over 2 years in the Explorys cohort and 3%/16% over 5 years in the PCMR cohort. Mean estimates of plasma NT-proBNP concentration indicative of half-maximum relative risk for events (EC50 values) at 2 and 5 years were 3730 pg/mL and 4199 pg/mL, respectively, values both close to the mean of 3880 pg/mL established for adults with HF. The plasma NT-proBNP concentration is suitable for estimating relative risk of mortality and heart transplantation in children with HF, independent of etiology and shows similar relations to clinical outcomes as in adults, indicating its likely value as a surrogate marker both for adult and pediatric HF.ClinicalTrials.gov Identifiers: NCT00005391 (May 26, 2000), NCT01873976 (June 10, 2013).

Novel architecture for gated recurrent unit autoencoder trained on time series from electronic health records enables detection of ICU patient subgroups.

Electronic health records (EHRs) are used in hospitals to store diagnoses, clinician notes, examinations, lab results, and interventions for each patient. Grouping patients into distinct subsets, for example, via clustering, may enable the discovery of unknown disease patterns or comorbidities, which could eventually lead to better treatment through personalized medicine. Patient data derived from EHRs is heterogeneous and temporally irregular. Therefore, traditional machine learning methods like PCA are ill-suited for analysis of EHR-derived patient data. We propose to address these issues with a new methodology based on training a gated recurrent unit (GRU) autoencoder directly on health record data. Our method learns a low-dimensional feature space by training on patient data time series, where the time of each data point is expressed explicitly. We use positional encodings for time, allowing our model to better handle the temporal irregularity of the data. We apply our method to data from the Medical Information Mart for Intensive Care (MIMIC-III). Using our data-derived feature space, we can cluster patients into groups representing major classes of disease patterns. Additionally, we show that our feature space exhibits a rich substructure at multiple scales.

Cohort design and natural language processing to reduce bias in electronic health records research.

Electronic health record (EHR) datasets are statistically powerful but are subject to ascertainment bias and missingness. Using the Mass General Brigham multi-institutional EHR, we approximated a community-based cohort by sampling patients receiving longitudinal primary care between 2001-2018 (Community Care Cohort Project [C3PO], n = 520,868). We utilized natural language processing (NLP) to recover vital signs from unstructured notes. We assessed the validity of C3PO by deploying established risk models for myocardial infarction/stroke and atrial fibrillation. We then compared C3PO to Convenience Samples including all individuals from the same EHR with complete data, but without a longitudinal primary care requirement. NLP reduced the missingness of vital signs by 31%. NLP-recovered vital signs were highly correlated with values derived from structured fields (Pearson r range 0.95-0.99). Atrial fibrillation and myocardial infarction/stroke incidence were lower and risk models were better calibrated in C3PO as opposed to the Convenience Samples (calibration error range for myocardial infarction/stroke: 0.012-0.030 in C3PO vs. 0.028-0.046 in Convenience Samples; calibration error for atrial fibrillation 0.028 in C3PO vs. 0.036 in Convenience Samples). Sampling patients receiving regular primary care and using NLP to recover missing data may reduce bias and maximize generalizability of EHR research.

Early prediction of decompensation (EPOD) score: Non-invasive determination of cirrhosis decompensation risk.

BACKGROUND & AIMS: Decompensation is a hallmark of disease progression in cirrhotic patients. Early detection of a phase transition from compensated cirrhosis to decompensation would enable targeted therapeutic interventions potentially extending life expectancy. This study aims to (a) identify the predictors of decompensation in a large, multicentric cohort of patients with compensated cirrhosis, (b) to build a reliable prognostic score for decompensation and (c) to evaluate the score in independent cohorts. METHODS: Decompensation was identified in electronic health records data from 6049 cirrhosis patients in the IBM Explorys database training cohort by diagnostic codes for variceal bleeding, encephalopathy, ascites, hepato-renal syndrome and/or jaundice. We identified predictors of clinical decompensation and developed a prognostic score using Cox regression analysis. The score was evaluated using the IBM Explorys database validation cohort (N = 17662), the Penn Medicine BioBank (N = 1326) and the UK Biobank (N = 317). RESULTS: The new Early Prediction of Decompensation (EPOD) score uses platelet count, albumin, and bilirubin concentration. It predicts decompensation during a 3-year follow-up in three validation cohorts with AUROCs of 0.69, 0.69 and 0.77, respectively, and outperforms the well-known MELD and Child-Pugh score in predicting decompensation. Furthermore, the EPOD score predicted the 3-year probability of decompensation. CONCLUSIONS: The EPOD score provides a prediction tool for the risk of decompensation in patients with cirrhosis that outperforms well-known cirrhosis scores. Since EPOD is based on three blood parameters, only, it provides maximal clinical feasibility at minimal costs.

Automated Design and Integration of Asset Administration Shells in Components of Industry 4.0.

One of the central concepts in the principles of Industry 4.0 relates to the methodology for designing and implementing the digital shell of the manufacturing process components. This concept, the Asset Administration Shell (AAS), embodies a systematically formed, standardized data envelope of a concrete component within Industry 4.0. The paper discusses the AAS in terms of its structure, its components, the sub-models that form a substantial part of the shell's content, and its communication protocols (Open Platform Communication-Unified Architecture (OPC UA) and MQTT) or SW interfaces enabling vertical and horizontal communication to involve other components and levels of management systems. Using a case study of a virtual assembly line that integrates AASs into the technological process, the authors present a comprehensive analysis centered on forming AASs for individual components. In the given context, the manual AAS creation mode exploiting framework-based automated generation, which forms the AAS via a configuration wizard, is assessed. Another outcome consists of the activation of a virtual assembly line connected to real AASs, a step that allows us verify the properties of the distributed manufacturing management. Moreover, a discrete event system was modeled for the case study, enabling the effective application of the Industry 4.0 solution.

NT-proBNP Qualifies as a Surrogate for Clinical End Points in Heart Failure.

N-terminal pro-B-type natriuretic peptide (NT-proBNP) is a well-established biomarker in heart failure (HF) but controversially discussed as a potential surrogate marker in HF trials. We analyzed the NT-proBNP/mortality relationship in real-world data (RWD) of 108,330 HF patients from the IBM Watson Health Explorys database and compared it with the NT-proBNP / clinical event end-point relationship in 20 clinical HF studies. With a hierarchical statistical model, we quantified the functional relationship and interstudy variability. To independently qualify the model, we predicted outcome hazard ratios in five phase III HF studies solely based on NT-proBNP measured early in the respective study. In RWD and clinical studies, the relationship between NT-proBNP and clinical outcome is well described by an Emax model. The NT-proBNP independent baseline risk (R0 , RWD/studies median (interstudy interquartile range): 5.5%/3.0% (1.7-4.9%)) is very low compared with the potential NT-proBNP-associated maximum risk (Rmax : 55.2%/79.4% (61.5-89.0%)). The NT-proBNP concentration associated with the half-maximal risk is comparable in RWD and across clinical studies (EC50 : 3,880/2,414 pg/mL (1,460-4,355 pg/mL)). Model-based predictions of phase III outcomes, relying on short-term NT-proBNP data only, match final trial results with comparable confidence intervals. Our analysis qualifies NT-proBNP as a surrogate for clinical outcome in HF trials. NT-proBNP levels after short treatment durations of less than 10 weeks quantitatively predict hazard ratios with confidence levels comparable to final trial readout. Early NT-proBNP measurement can therefore enable shorter and smaller but still reliable HF trials.

New Approaches to Implementing the SmartJacket into Industry 4.0 ‡.

The paper discusses the possibilities of incorporating sensors and indicators into the environment of an Industry 4.0 digital factory. The concept of Industry 4.0 (I4.0) is characterized via a brief description of the RAMI 4.0 and I4.0 component model. In this context, the article outlines the structure of an I4.0 production component, interpreting such an item as a body integrating the asset and its electronic form, namely, the Asset Administration Shell (AAS). The formation of the AAS sub-models from the perspectives of identification, communication, configuration, safety, and condition monitoring is also described to complete the main analysis. Importantly, the authors utilize concrete use cases to demonstrate the roles of the given I4.0 component model and relevant SW technologies in creating the AAS. In this context, the use cases embody applications where an operator wearing a SmartJacket equipped with sensors and indicators ensures systematic data collection by passing through the manufacturing process. The set of collected information then enables the operator and the system server to monitor and intervene in the production cycle. The advantages and disadvantages of the individual scenarios are summarized to support relevant analysis of the entire problem.

Machine Learning based Human Gait Segmentation with Wearable Sensor Platform.

Supervised and unsupervised machine learning algorithms were explored for gait segmentation using wearable sensor platform. Multiple wearable sensors modules were placed at key locations: Four Inertial Measurement Units (IMUs) were attached to the thigh and shank of each leg and a plantar pressure measuring foot insoles were implanted in the shoes. The gait data has been collected from 10 people wirelessly via TCI-IP protocol, which is later anonymized. Further, the Ranchos Los Amigos (RLA) gait nomenclature-based data preprocessing and peak/valley detector based annotation steps are performed on the acquired data followed by implementation of machine learning techniques on the labeled datasets. The methods explored for phase and sub-phase classification includes the Unsupervised methods such as K-Means clustering and supervised methods like the Support Vector Machine (SVM) and Artificial Neural Network (ANN).

Deep Learning based Gait Abnormality Detection using Wearable Sensor System.

Gait is an extraordinary complex function of human body that involves the activation of entire visceral nervous system, making human gait definite to various functional abnormalities. Diagnosis and treatment of such disorders prior to their development can be achieved through integration of modern technologies with state-of-the-art developed methods. Modern machine learning techniques have outperformed and complemented the use of conventional statistical methods in bio-medical systems. In this research a wearable sensor system is presented, which combines plantar pressure measurement unit and Inertial Measurement Units (IMU's) integrated with a stacked Long short-term memory (LSTM) model to detect human gait abnormalities that are prone to the risk of fall. The computed metrics and gait parameters show significant differences between normal and abnormal gait patterns. Three specific abnormalities involving Hemiplegic, Parkinsonian and Sensory-Ataxic gaits are simulated to validate the proposed model and show promising results. The proposed research aims to demonstrate how advanced technologies can be used in gait diagnosis and treatment assistant systems.

Structural and mechanistic insights into the oxy form of tyrosinase from molecular dynamics simulations.

The first, long time scale (16-ns) ligand field molecular dynamics (LFMD) simulations of the oxy form of tyrosinase are reported. The calculations use our existing type 3 copper force field for the peroxido-bridged [Cu(2)O(2)](2+) unit which is here translated from MMFF into the AMBER format together with a new charge scheme. The protein secondary and tertiary structures are not significantly altered by removing the 'caddie' protein, ORF378, which must be bound to tyrosinase before crystals will grow. A comprehensive principal component analysis of the Cartesian coordinates from the final 8 ns shows that the protein backbone is relatively rigid. However, the significant butterfly fold of the [Cu(2)O(2)](2+) moiety observed in the X-ray structure, presumably due to the caddie protein tyrosine at the active site, is absent in the simulations. LFMD gives a clear and persistent distinction between equatorial and axial Cu-N distances, with the latter about 0.2 A longer and remaining syn to each other. However, the two coordination spheres display important differences. LFMD simulations of the symmetric model complex [mu-eta(2):mu(2)-O(2){Cu(Meim)(3)}(2)](2+) (Meim is 5-methyl-1H-imidazole) provide a mechanism for syn-anti interchange of axial ligands which suggests, in combination with the old experimental X-ray data, the new LFMD simulations and traditional coordination chemistry arguments, that His(54) on Cu(A) is 'insipiently axial' and that a combination of a butterfly distortion of the [Cu(2)O(2)](2+) group and a rotation of the Cu(A)(His)(3) moiety converts the vacant, initially axial, binding site on Cu(A) into a much more favourable equatorial site.

On the performance of ligand field molecular mechanics for model complexes containing the peroxido-bridged [Cu2O2]2+ center.

The ability of ligand field molecular mechanics (LFMM) to model accurately the structures and relative conformer energies of complexes containing the [Cu2O2](2+) unit found in oxidized copper type 3 (T3) enzymes is investigated. The consequences of ignoring the coupling between the metal centers are analyzed and predicted to be unimportant with respect to computing molecular geometries. Angular overlap model (AOM) parameters for the peroxido bridge in [Cu2O2](2+) are derived on the basis of the mononuclear model compound [(NH3)3CuO2] for which good ligand field and density functional theory (DFT) calculations are also possible. Metal-ligand pi-bonding parameters are shown to play an important role with the in-plane AOM pi-bonding parameter value being significantly larger than that for the out-of-plane parameter. The LFMM treatment is then extended to the model dinuclear species [(H3N)3CuO2Cu(NH3)3](2+). The planarity of the [Cu2O2](2+) moeity is implicitly obtained by defining the directions of the local Cu-O pi-bonding interactions with respect to the other copper atom, rather than the other oxygen. In conjunction with the force field parameters based on the Merck molecular force field, the model, as implemented in our DommiMOE program (Deeth, R. J. ; Fey, N. ; Williams-Hubbard, B. J. J. Comput. Chem. 2005, 26, 123- 130), is applied to a set of crystallographically characterized small-molecule mimics of the T3 active site. Extensive LFMM conformational searches are carried out for these compounds, and the quality of the LFMM potential energy hypersurface is assessed by comparison with results using DFT. We find that the description of the geometries does not in fact suffer from the neglect of explicit coupling between the metal centers. Moreover, the structures and relative energies obtained by the LFMM conformational searches agree well with both experiment and the DFT values for all systems except one where the LFMM structure which is in best agreement with experiment is about 10 kcal mol(-1) higher than the lowest energy conformer. However, this discrepancy is traced to generic shortcomings in the "organic" force field rather than the LFMM.

Pseudohelical and helical primary structures of 1,2-spiroannelated four- and five-membered rings: syntheses and chiroptical properties.

The pseudohelical hydrocarbons (R)-6, (S)-7, and (R)-8 and the helical hydrocarbon (P)-9, formally derived from the helical hydrocarbon (P)-4 by stepwise replacement of each of the four-membered rings by a five-membered ring, have been prepared. Their optical rotations vary systematically, both in magnitude and sign. Of the extremes, (P)-4 represents the usual case of a right-handed dextrorotatory helix, while (P)-9 represents the unusual case of a right-handed levorotatory helix. To rationalize these facts, DFT calculations of the rotatory power of (P)-helices of three-, four-, and five-membered rings have been performed. The results show a very good agreement with the experimental data for the rigid helices of three-membered rings and always show the correct sign and order of magnitude for the flexible helices of four- and five-membered rings for which Boltzmann-averaged optical rotations of up to six conformers had to be used. Within the conformers of the latter, a set of large dihedral angles for the bonds of the inner sphere correspond to a high specific rotation, and a set of small dihedral angles correspond to a low specific rotation. As a consequence, the Boltzmann-averaged values markedly depend on the geometry and weight of the conformers involved.

Metallosupramolecular chemistry with bis(benzene-o-dithiolato) ligands.

The bis(benzene-o-dithiol) ligands H(4)-1, H(4)-2, and H(4)-3 react with [Ti(OC(2)H(5))(4)] to give dinuclear triple-stranded helicates [Ti(2)L(3)](4)(-) (L = 1(4)(-), 2(4)(-), 3(4)(-)). NMR spectroscopic investigations revealed that the complex anions possess C(3) symmetry in solution. A crystal structure analysis for (PNP)(4)[Ti(2)(2)(3)] ((PNP)(4)[14]) confirmed the C(3) symmetry for the complex anion in the solid state. The complex anion in Li(PNP)(3)[Ti(2)(1)(3)] (Li(PNP)(3)[13]) does not exhibit C(3) symmetry in the solid state due to the formation of polymeric chains of lithium bridged complex anions. Complexes [13](4)(-) and [14](4)(-) were obtained as racemic mixtures of the Delta,Delta and Lambda,Lambda isomers. In contrast to that, complex (PNP)(4)[Ti(2)(3)(3)] ((PNP)(4)[15]) with the enantiomerically pure chiral ligand 3(4)(-) shows a strong Cotton effect in the CD spectrum, indicating that the chirality of the ligands leads to the formation of chiral metal centers. The o-phenylene diamine bridged bis(benzene-o-dithiol) ligand H(4)-4 reacts with Ti(4+) to give the dinuclear double-stranded complex Li(2)[Ti(2)(4)(2)(mu-OCH(3))(2)] containing two bridging methoxy ligands between the metal centers. The crystal structure analysis and the (1)H NMR spectrum of (Ph(4)As)(2)[Ti(2)(4)(2)(mu-OCH(3))(2)] ((Ph(4)As)(2)[(16]) reveal C(2) symmetry for the anion [Ti(2)(4)(2)(mu-OCH(3))(2)](2)(-). For a comparative study the dicatechol ligand H(4)-5, containing the same o-phenylene diamine bridging group as the bis(benzene-o-dithiol) ligands H(4)-4, was prepared and reacted with [TiO(acac)(2)] to give the dinuclear complex anion [Ti(2)(5)(2)(mu-OCH(3))(2)](2)(-). The molecular structure of (PNP)(2)[Ti(2)(5)(2)(mu-OCH(3))(2)] ((PNP)(2)[17]) contains a complex anion which is similar to [16](2)(-), with the exception that strong N-H...O hydrogen bonds are formed in complex anion [17](2)(-), while N-H...S hydrogen bonds are absent in complex anion [16](2)(-).

Weak intermolecular interactions calculated with diffusion Monte Carlo.

The performance of fixed node diffusion Monte Carlo (FNDMC) for weakly interacting molecules is investigated. The effect of Gaussian basis sets on the asymptotic description of the molecular orbitals which is crucial for a successful importance sampling is analyzed for the example of the hydrogen atom. We find that accurate reference binding energies of the water, the ammonia, and the T-shaped as well as the parallel-displaced benzene dimer are correctly reproduced by FNDMC. The binding energies for the benzene dimers are -3.00(0.38) and -3.58(0.38) kcal/mol, respectively. The description of the methane dimer which has the smallest binding energy and a quite large intermolecular distance requires a more flexible basis set of diffuse quadruple-zeta quality in order to prevent sampling errors.

Performance of diffusion Monte Carlo for the first dissociation energies of transition metal carbonyls.

Fixed node diffusion Monte Carlo (FNDMC) calculations are carried out for the first ligand dissociation energies of the prototype transition metal carbonyls Cr(CO)6, Fe(CO)5, Ni(CO)4, and Fe(CO)4N2. Since Hartree-Fock theory performs particularly badly for these type of compounds they are difficult to treat with conventional ab initio methods. We find that a Kohn-Sham determinant from a standard density functional provides a balanced description of the fermionic nodal hyper surfaces of all compounds involved in the dissociation reaction. With one exception, the experimental dissociation enthalpies are reproduced by FNDMC within the statistical accuracy of the method.

Function block applications in control systems based on IEC 61804.

A variety of fieldbus technologies and digital fieldbus devices have been introduced within the process industries over the last ten years. There has been a gradual acceptance of the fact that a variety of communication technologies are needed to fully address the application requirements of a manufacturing facility. However, engineers responsible for the specification, engineering, and implementation of control systems require that a common interface and functionality be provided in the control system. This capability should be independent of the underlying fieldbus technology or manufacturer of the fieldbus device. The draft IEC 61804 standard defines how a control system can be structured to provide this flexibility in the utilization of fieldbus technology. In this paper, we discuss how a consistent function block capability may be provided for all fieldbus technology utilized in a control system. Examples will be given of how this standard has been applied in modern control systems to give a consistent interface to Foundation Fieldbus and PROFIBUS. Some detail will be presented on the standard means that is defined for manufacturers to describe function block capability of a field device. An analysis is given of the impact and benefit that the IEC 61804 standard will have on the process industry and on manufacturers of control systems.