Pairs Of Data Points Research Articles

Automated mapping of electronic data capture fields to SDTM.

The goal of this work is to reduce the amount of manual work required to go from data capture to regulatory submission. It will be shown that the use of Siamese networks will allow for the generation of embeddings that can be used by traditional machine learning classifiers to perform the classification at much higher levels of accuracy than standard approaches. Siamese networks are a method for training data embeddings such that data within the same class are closer with respect to a given distance metric than they are to data points in another class. Because they are designed to learn similarity within pairs of data points, they work well in situations where the number of classes is relatively large compared to the number of training samples. In this work, we will show that embeddings generated via a Siamese network from metadata associated with electronic data capture forms can be used to predict the associated SDTM field. With a relatively simple network coupled with a basic classification algorithm, the proposed method can achieve accuracies greater than 90%, which is significantly higher than what has been achieved with traditional methods, with many of the inaccurate mappings due to a lack of training data. In many cases, there is a 15% increase in accuracy vs. more traditional methods. Leveraging Siamese networks, it is possible to generate embeddings that efficiently represent data fields in a lower dimensional space. This allows the creation of a system that can automatically map between data schemas at high levels of accuracy. Such systems represent the first step in automating one of the many labor-intensive data management tasks associated with clinical trials.

Correlation of temperature-sensing microchip and rectal temperature measurements in cats.

Rectal temperature (RT) is the reference standard for clinical evaluation of body temperature in mammals. However, the use of a rectal thermometer to measure temperature can cause stress and other problems, especially in cats. There is a need for clinical techniques that reduce both stress and defensive behavior as part of the provision of better medical care. Subcutaneous temperature-sensing identification microchips fulfil the current legal requirements and provide a reading of subcutaneous temperature (MT). The clinical study tried to determine whether there is agreement between MT and RT in normal (n = 58), hospitalized (n = 26) and sedated/anesthetized (n = 36) cats. Three measurements were taken using both methods (MT and RT) in each cat. Correlation between MT and RT, and differences between MT and RT, were estimated for pairs of data-points from the same individual, and all data pairs in each group were considered overall. There was a strong positive correlation between MT and RT (r = 0.7 to 1.0) (p < 0.0005). The mean differences (d) were always negative and although statistically significant, these d values are likely of no biological importance. The overall d was ‑0.1°C in normal cats (p < 0.0005), -0.1°C in hospitalized cats (p = 0.001) and -0.1°C in sedated/anesthetized cats (p = 0.001). The limits of agreement between MT and RT appear narrow enough for MT to be acceptable estimate of RT. The overall limits of agreement (95%) were ‑0.71°C and 0.53°C (in normal cats); ‑0.51°C and 0.34°C (in hospitalized cats) and ‑0.60°C and 0.42°C (in sedated/anesthetized cats). MT may provide a good alternative to RT measurement in cats. However, this study was mostly performed in animals that were normothermic. Therefore, further studies in larger groups of cats under different conditions are needed to compare trends and assess variation with time.

An entropy-based weighted dissimilarity metric for numerical data clustering using the distribution of intra feature differences

Suitable selection of a proximity measure is one of the fundamental requirements of clustering. With conventional (dis)similarity measures, many clustering algorithms do not yield satisfactory results on complex high-dimensional datasets. The problem lies with widely varying distributions along each feature which existing (dis)similarity measures fail to capture. In this work, we study the distribution of all-pair absolute distances in some standard real datasets over the feature space and observe that most of them have values near to zero. The frequency of data pairs decreases with the increasing value of distance which suggests an exponential distribution of values along features. The exponential decay rate constant, termed as characteristic length indicates the inhomogeneity in a feature and therefore, we use it as a weighting factor across attributes. The dissimilarity for a pair of data points is computed by considering the weights of each attribute along with a continuum adaptation of Boltzmann’s notion of entropy which uses feature-wise absolute differences as input. We prove that the proposed measure is a metric. For experimental analysis, we compare different proximity measures with the proposed, in terms of clustering results. The combination of feature wise characteristic length and the continuous version of Boltzmann’s entropy proves its excellence in terms of clustering results on diversified synthetic, real, and gene expression datasets.

Multi-view clustering with Laplacian rank constraint based on symmetric and nonnegative low-rank representation

Multi-view spectral clustering has attracted more and more attention due to its promising advantages in data clustering, and many related methods have been proposed, among which the method based on low-rank representation have become increasingly popular. However, the similarity matrices learned by the existing methods inevitably contain some negative entries and are generally nonsymmetric. In order to guarantee weight consistency for each pair of data points and highly correlated data points to be together, we propose a new multi-view spectral clustering model based on the symmetric and nonnegative low-rank representation. This model can learn the view-specific consistent similarity matrix and also capture the local structure and discriminative information among data by imposing symmetric and nonnegative constraints on the representation matrix. By employing the compatible and complementary information from multi-view data points, it can also learn the consensus similarity matrix for all views and perform the spectral clustering simultaneously. By the augmented Lagrangian multiplier method, an alternating iterative optimization algorithm is designed for solving this model and its computational complexity and convergence is analyzed. Comprehensive experiments on multiple multi-view data sets also verify the effectiveness and superiority of the proposed approach.

Clustering performance analysis using a new correlation-based cluster validity index

There are various cluster validity indices used for evaluating clustering results. One of the main objectives of using these indices is to seek the optimal unknown number of clusters. Some indices work well for clusters with different densities, sizes, and shapes. Yet, one shared weakness of those validity indices is that they often provide only one optimal number of clusters. That number is unknown in real-world problems, and there might be more than one possible option. We develop a new cluster validity index based on a correlation between an actual distance between a pair of data points and a centroid distance of clusters that the two points occupy. Our proposed index constantly yields several local peaks and overcomes the previously stated weakness. Several experiments in different scenarios, including UCI real-world data sets, have been conducted to compare the proposed validity index with several well-known ones.

Fast and Robust Unsupervised Dimensionality Reduction with Adaptive Bipartite Graphs

Recent unsupervised dimension reduction algorithms use similarity graphs between data point pairs to preserve local structure while reducing dimension. However, the time complexity of these methods is proportional to the square of the number of samples, which limits their application to large-scale datasets. Moreover, the square Euclidean calculation criterion between sample point pairs will magnify the bad influence of outliers on the graph. In addition, these methods only preserve the local structure while losing other important structural information. To this end, we propose a fast adaptive unsupervised projection model termed Fast and Robust Unsupervised Dimensionality Reduction with Adaptive Bipartite Graph (FRUDR-ABG), which uses a few anchor points and sample points to build a bipartite graph to preserve the local geometric structure of the data to reduce the running time and improve efficiency. We propose a criterion based on the l2,1 norm to calculate the distance between anchor points and data points to reduce the negative influence of outliers on graph construction. A practical strategy is also proposed to realize joint learning of global and local structures. According to the characteristics of graph construction and dimensionality reduction adaptive learning in the algorithm, we design an iterative reweighting method to solve the model. Experimental results on several benchmark datasets show that FRUDR-ABG has higher efficiency and recognition performance than existing unsupervised dimensionality reduction methods.

Capacity estimation for lithium-ion battery via a novel health indicator extracted from partial constant voltage charging curve

This paper develops a capacity estimation method via a novel health indicator (HI) extracted from partial constant voltage (CV) charging curve. Firstly, the CV charging curve is analyzed based on the Thevenin model, whose results indicate that the CV charging curves under different battery aging states can be theoretically described by the same functional form. Secondly, with CV charging curve analysis results, regarding the CV charging curve at the first cycle as original curve, and those at other cycles as fitted curves, the coefficient of determination between those two kinds of curves is computed through the early-stage CV charging data and innovatively taken as HI for further capacity estimation. Meanwhile, considering that the early-stage CV charging data may still be insufficient in real application, a simple subsequent current prediction method is proposed to ease HI extraction in the case of insufficient data based on the first data point of CV charging curve. Thirdly, a moving window-based HI processing method is developed to mitigate the negative influence of the possible current measurement disturbances on capacity estimation accuracy. Finally, the proposed method is verified through Underwriters Laboratories Inc. - Purdue University battery aging dataset, whose results demonstrate that both mean absolute error and root mean squared error of capacity estimation results achieved by the model established by only 12 pairs of data points in the premise of appropriately selected original curve can be roughly controlled less than 1% and 1.5% in the case of sufficient data and insufficient data, respectively.

A novel cluster validity index based on augmented non-shared nearest neighbors

The true cluster number of the dataset in practical applications is rarely known in advance. Therefore, it is necessary to use a cluster validity index to evaluate the clustering results and determine the optimal cluster number. However, the performance of existing cluster validity indices is vulnerable to various factors such as cluster shape and density. To solve the above issues, this paper proposes a new cluster validity index based on augmented non-shared nearest neighbors (ANCV). The ANCV index is based on the following principles: (1) Within-cluster compactness can be measured by the distance between the pairs of data points with fewer shared nearest neighbors. (2) The distances between the pairs of data points at the intersection of clusters can be used to estimate the between-cluster separation. On this basis, the above point pairs are further extended to their augmented non-shared nearest neighbors, thereby forming small clusters. Then, the average distance within and between these clusters is calculated respectively to estimate the within-cluster compactness and between-cluster separation. Finally, the optimal number of clusters is determined by the difference between the between-cluster separation and the within-cluster compactness. Experimental results on both 12 two-dimensional synthetic datasets and 10 real datasets from UCI have shown that the ANCV index performs the best among all compared indices.

Domain Adaptation Principal Component Analysis: Base Linear Method for Learning with Out-of-Distribution Data

Domain adaptation is a popular paradigm in modern machine learning which aims at tackling the problem of divergence (or shift) between the labeled training and validation datasets (source domain) and a potentially large unlabeled dataset (target domain). The task is to embed both datasets into a common space in which the source dataset is informative for training while the divergence between source and target is minimized. The most popular domain adaptation solutions are based on training neural networks that combine classification and adversarial learning modules, frequently making them both data-hungry and difficult to train. We present a method called Domain Adaptation Principal Component Analysis (DAPCA) that identifies a linear reduced data representation useful for solving the domain adaptation task. DAPCA algorithm introduces positive and negative weights between pairs of data points, and generalizes the supervised extension of principal component analysis. DAPCA is an iterative algorithm that solves a simple quadratic optimization problem at each iteration. The convergence of the algorithm is guaranteed, and the number of iterations is small in practice. We validate the suggested algorithm on previously proposed benchmarks for solving the domain adaptation task. We also show the benefit of using DAPCA in analyzing single-cell omics datasets in biomedical applications. Overall, DAPCA can serve as a practical preprocessing step in many machine learning applications leading to reduced dataset representations, taking into account possible divergence between source and target domains.

Research on incremental clustering algorithm for big data

Abstract As the scale of data becomes larger and larger, clustering processing, a key step in data mining, has important practical significance. Aiming at the problems of time consumption and high clustering errors when the current clustering algorithms deal with massive and dynamic big data, an incremental clustering algorithm is proposed by taking big data as the research object. By exploring the attribute characteristics of big data, four characteristics such as scale, diversity, high speed and value are summarised. For large-scale data streams that have multiple attributes and are acquired one by one, optimise the setting method of the K-means clustering algorithm category centre point, combine the K-means clustering algorithm and the Kalman filter algorithm and measure the distance between data point pairs. Instead of Mahalanobis distance, an incremental clustering algorithm suitable for big data is constructed. Five data sets are selected to carry out example analysis. The results of the algorithm are verified by the algorithm. The proposed algorithm has obvious advantages in the incremental clustering effect of big data. At the same time, it also has efficient and stable computing performance, which meets the expected design requirements and goals.

Estimating hydrogen absorption energy on different metal hydrides using Gaussian process regression approach

Hydrogen is a promising alternative energy source due to its significantly high energy density. Also, hydrogen can be transformed into electricity in energy systems such as fuel cells. The transition toward hydrogen-consuming applications requires a hydrogen storage method that comes with pack hydrogen with high density. Among diverse methods, absorbing hydrogen on host metal is applicable at room temperature and pressure, which does not provide any safety concerns. In this regard, AB2 metal hydride with potentially high hydrogen density is selected as an appropriate host. Machine learning techniques have been applied to establish a relationship on the effect of the chemical composition of these hosts on hydrogen storage. For this purpose, a data bank of 314 data point pairs was used. In this assessment, the different A-site and B-site elements were used as the input variables, while the hydrogen absorption energy resulted in the output. A robust Gaussian process regression (GPR) approach with four kernel functions is proposed to predict the hydrogen absorption energy based on the inputs. All the GPR models' performance was quite excellent; notably, GPR with Exponential kernel function showed the highest preciseness with R2, MRE, MSE, RMSE, and STD of 0.969, 2.291%, 3.909, 2.501, and 1.878, respectively. Additionally, the sensitivity of analysis indicated that ZR, Ti, and Cr are the most demining elements in this system.

Narrow laser-linewidth measurement using short delay self-heterodyne interferometry.

Delayed self-heterodyne interferometry is a commonly used technique for the measurement of laser linewidth. It typically requires the use of a very long delay fiber when measuring narrow linewidth (especially linewidths in the kHz-range) lasers. The use of long fibers can result in system losses and the introduction of 1/f noise that causes spectral line broadening. In this paper, we present a calculation method for processing the output of a delayed self-heterodyne setup using a short length of delay fiber, to determine laser linewidth. The method makes use of pairs of data points (corresponding to adjacent maxima and/or minima) in the signal generated from the self-heterodyne setup to determine the laser linewidth. Here, the power ratio or amplitude difference of the signal at these data points is of importance. One of the key benefits of this method is that it avoids 1/f noise which would otherwise be introduced into the measurement through the application of long fibers. The experimental results highlight that the method has a high calculation accuracy. Furthermore, the capacity for the method to utilize different pairs of data points in the self-heterodyne output to determine the laser linewidth, imparts a high degree of flexibility and usability to the technique when applied to real-world measurements.

A computational cardiopulmonary physiology simulator accurately predicts individual patient responses to changes in mechanical ventilator settings.

We present new results validating the capability of a high-fidelity computational simulator to accurately predict the responses of individual patients with acute respiratory distress syndrome to changes in mechanical ventilator settings. 26 pairs of data-points comprising arterial blood gasses collected before and after changes in inspiratory pressure, PEEP, FiO2, and I:E ratio from six mechanically ventilated patients were used for this study. Parallelized global optimization algorithms running on a high-performance computing cluster were used to match the simulator to each initial data point. Mean absolute percentage errors between the simulator predicted values of PaO2 and PaCO2 and the patient data after changing ventilator parameters were 10.3% and 12.6%, respectively. Decreasing the complexity of the simulator by reducing the number of independent alveolar compartments reduced the accuracy of its predictions. Clinical Relevance- These results provide further evidence that our computational simulator can accurately reproduce patient responses to mechanical ventilation, highlighting its usefulness as a clinical research tool.

Soil nitrogen status can be improved through no-tillage adoption particularly in the surface soil layer: A global meta-analysis

United Nations ‘Sustainable Nitrogen Management’ (UNEP/EA.4/L.16) resolution calls for adopting best practices for recovery and recycling of N. Bulk of the soil N stock is in soil organic matter (SOM) form. The stabilization of SOM following conversion to no-till practice will, therefore, regulate N flux and storage in soils. To realize the no-tillage (NT) effect on soil N, a global meta-analysis was conducted with 2,268 pairs of data points from 327 peer-reviewed articles. Increases of 21 and 16% in total N concentration and N stock, respectively, were noted under NT compared to conventional tillage (CT) in the surface soil layer (0–10 cm). However, no such changes were accounted for down the profile. Continental and temperate climates recorded significant gains in total N in NT while the effect was negligible in tropical and dry climates. Soil N increased in NT irrespective of soil texture; coarser the texture, marginally higher was the gain. An incremental rise in soil N was evident with an increase in the duration of NT adoption. The cereal-cereal and cereal-legume rotations had marginal advantages with NT over other cropping systems. The surface 0–10 cm layer recorded the largest N stock under NT, which remained similar down the depth. A decrease in soil pH was noted with conversion to NT practice, but soil electrical conductivity was comparable with CT. No-tillage favoured the soil N and as a consequence, fertilizer-N use under no-till condition, must take into account of the potential N release from SOM for achieving higher N-use efficiency.

Hashing-based affinity matrix for dominant set clustering

Dominant set clustering has been widely used to solve a variety of problems such as image segmentation, video analysis, and image retrieval. However, the key problem of it is the need of a given affinity matrix to provide similarity for every pair of data points. The affinity matrix is either user-given or computed using a pairwise computation, which is impractical for big data situations. Therefore, this work proposes a semi-supervised hashing-based affinity matrix computation method (HAM) for dominant set clustering (HAM-DSC) for cases of semi-supervised clustering where labels are available to a portion of data. The HAM computes affinity matrix efficiently based on Hamming distance between learned hash codes, which preserve semantic similarities effectively. To deal with unsupervised clustering problems, an unsupervised extension of the HAM-DSC is also given in this work. Experimental results on 7 real-world datasets show that the HAM-DSC yields an overall better performance together with lower time costs compared to dominant set clustering using other affinity matrix computation methods.

Active Classification With Uncertainty Comparison Queries.

Noisy pairwise comparison feedback has been incorporated to improve the overall query complexity of interactively learning binary classifiers. The positivity comparison oracle is extensively used to provide feedback on which is more likely to be positive in a pair of data points. Because it is impossible to determine accurate labels using this oracle alone without knowing the classification threshold, existing methods still rely on the traditional explicit labeling oracle, which explicitly answers the label given a data point. The current method conducts sorting on all data points and uses explicit labeling oracle to find the classification threshold. However, it has two drawbacks: (1) it needs unnecessary sorting for label inference and (2) it naively adapts quick sort to noisy feedback. In order to avoid these inefficiencies and acquire information of the classification threshold at the same time, we propose a new pairwise comparison oracle concerning uncertainties. This oracle answers which one has higher uncertainty given a pair of data points. We then propose an efficient adaptive labeling algorithm to take advantage of the proposed oracle. In addition, we address the situation where the labeling budget is insufficient compared to the data set size. Furthermore, we confirm the feasibility of the proposed oracle and the performance of the proposed algorithm theoretically and empirically.

Thermal analysis of Alzheimer’s disease prediction using random forest classification model

Missing and irregularly sampled data points are common in time-dependent data collected in Alzheimer's disease studies. As a result, time-series methods based on regular sampling cannot be applied directly to data without first undergoing pre-processing. In this paper developed to learn the relationship between pairs of data points at different times, we implement random forest model separations in time the input vector contains a summary of the time series history. Due to similar brain patterns and pixel intensities, the diagnosis of AD images to feature representation for categorization. Machine Learning approaches can learn things like representations from data. A 4-way classifier is used in this study to distinguish between Healthy Control (HC) and Mild Alzheimer's disease (AD), Very Mild AD, and Moderate AD patients. Experiments are carried out using the OASIS dataset, and the results are used to classify diseases into multiple categories. The Random Forest Classification Model was used to do a thermal analysis of Alzheimer's disease prediction. The random forest classifier models have been shown to consistently outperform their equivalents. The improved accuracy of tweaking the parameters of random forest classifiers demonstrated in this study work argues for its use in the classification of stages of Alzheimer's disease.

A New Fuzzy Cluster Validity Index for Hyperellipsoid or Hyperspherical Shape Close Clusters With Distant Centroids

Determining the correct number of clusters is essential for efficient clustering and cluster validity indices are widely used for the same. Generally, the effectiveness of a cluster validity index relies on two factors: first, separation, defined by the distance between a pair of cluster centroids or a pair of data points belonging to different clusters and second, compactness, which is determined in terms of the distance between a data point and a centroid or between a pair of data points belonging to the same cluster. However, the existing cluster validity indices for centroid-based clustering are unreliable when the clusters are too close, but corresponding centroids are distant. To mitigate this, a new cluster validity index, Saraswat-and-Mittal index, has been proposed in this article for hyperellipsoid or hyperspherical shape close clusters with distant centroids, generated by fuzzy c-means. The proposed index computes compactness in terms of the distance between data points and corresponding centroids, whereas the distance between data points of disjoint clusters defines separation. These parameters benefit the proposed index in the analysis of close clusters with distinct centroids efficiently. The performance of the proposed index is validated against ten state-of-the-art cluster validity indices on artificial, UCI, and image datasets, clustered by the fuzzy c-means.

Global meta-analysis suggests that no-tillage favourably changes soil structure and porosity

Role of soil to meet global food security, sustainable intensification and food nutritional quality has got renewed attention with a larger focus on soil physical condition. No-tillage (NT) practice can essentially contribute to develop a sustainable, low carbon and resource efficient agriculture, and encourage the use of crop residues for added soil benefits. Soil aggregation and pore size distribution, two most important soil physical factors controlling the mass and energy transport processes within the soil and between soil and environment, were evaluated under the NT through a global meta-analysis of 5065 pairs of data points from 419 peer-reviewed studies. Compared to conventional tillage (CT), NT increased mean weight diameter of aggregates, water stable aggregates, and macroaggregates by averages (0–30 cm) of 25, 10 and 22%, respectively, although predominantly in 0–10 and/or 10–20 cm layers, with an accompanying reduction in microaggregates. A small but significant 3% decrease in total porosity, a large reduction (20–32%) in macroporosity and a moderate increase (4–7%) in microporosity were realized under NT up to 20 cm soil depth. Bulk density remained stable, although a very large decrease (70% change over CT) in saturated hydraulic conductivity was recorded in 10–20 and >30 cm soil layers. Years of adoption of NT had an additive effect on mean weight diameter and macroaggregates, and the total and macroporosity. Increase in latitudes favoured soil aggregation and micropore volume under NT, while clay content was unfavourable to macro- and water stable aggregate contents. Improvement in structure and water retention properties relate to long-term sustainable development of soils by following no-till practice, which has far-reaching implications beyond the boundaries of agronomy.

Negative drift of sedation depth in critically ill patients receiving constant minimum alveolar concentration of isoflurane, sevoflurane, or desflurane: a randomized controlled trial

BackgroundIntensive care unit (ICU) physicians have extended the minimum alveolar concentration (MAC) to deliver and monitor long-term volatile sedation in critically ill patients. There is limited evidence of MAC’s reliability in controlling sedation depth in this setting. We hypothesized that sedation depth, measured by the electroencephalography (EEG)-derived Narcotrend-Index (burst-suppression N_Index 0—awake N_Index 100), might drift downward over time despite constant MAC values.MethodsThis prospective single-centre randomized clinical study was conducted at a University Hospital Surgical Intensive Care Unit and included consecutive, postoperative ICU patients fulfilling the inclusion criteria. Patients were randomly assigned to receive uninterrupted inhalational sedation with isoflurane, sevoflurane, or desflurane. The end-expiratory concentration of the anaesthetics and the EEG-derived index were measured continuously in time-stamped pairs. Sedation depth was also monitored using Richmond-Agitation-Sedation-Scale (RASS). The paired t-test and linear models (bootstrapped or multilevel) have been employed to analyze MAC, N_Index and RASS across the three groups.ResultsThirty patients were recruited (female/male: 10/20, age 64 ± 11, Simplified Acute Physiology Score II 30 ± 10). In the first 24 h, 21.208 pairs of data points (N_Index and MAC) were recorded. The median MAC of 0.58 ± 0.06 remained stable over the sedation time in all three groups. The t-test indicated in the isoflurane and sevoflurane groups a significant drop in RASS and EEG-derived N_Index in the first versus last two sedation hours. We applied a multilevel linear model on the entire longitudinal data, nested per patient, which produced the formula N_Index = 43 − 0.7·h (R2 = 0.76), showing a strong negative correlation between sedation’s duration and the N_Index. Bootstrapped linear models applied for each sedation group produced: N_Index of 43–0.9, 45–0.8, and 43–0.4·h for isoflurane, sevoflurane, and desflurane, respectively. The regression coefficient for desflurane was almost half of those for isoflurane and sevoflurane, indicating a less pronounced time-effect in this group.ConclusionsMaintaining constant MAC does not guarantee stable sedation depth. Thus, the patients necessitate frequent clinical assessments or, when unfeasible, continuous EEG monitoring. The differences across different volatile anaesthetics regarding their time-dependent negative drift requires further exploration.Trial registration: NCT03860129.