Noise Level Of Data Research Articles

A novel deep domain adaptation method based on three-way decisions for identifying autistic patients

Combining machine learning with functional magnetic resonance imaging (fMRI) technology to build an effective autism recognition model has become a research hotspot. However, the performance of existing machine learning models is highly dependent on large-scale labelled data, and the cost of fMRI data acquisition and labelling limits the application of models in autism recognition. Although the availability of multisite data increases the sample size, the resolution, image contrast, and noise level of fMRI data acquired at different sites vary considerably due to the different scanning devices or scanning protocols used. The heterogeneity of fMRI data between sites leads to limited generalisation of the trained models. To tackle this problem, we propose a deep domain adaptation approach based on three-way decisions. First, we introduce a pseudolabel optimisation method based on three-way decisions, which fully considers the structural information of the target domain and makes secondary decisions on the uncertain objects in the target domain, thus improving the accuracy of the pseudolabel in the target domain. Then, we construct a new autism recognition network using a stacked autoencoder and combine the proposed pseudolabel optimisation model to adapt the domain shift. Finally, the self-training strategy is used to retrain the network after domain adaptation to improve the generalisability of the model to the target domain. Experimental results on several cross-domain autism recognition tasks show the effectiveness of the proposed method.

An improved spectral clustering method for accurate detection of brain resting-state networks

This paper proposes a data-driven analysis method to accurately partition large-scale resting-state functional brain networks from fMRI data. The method is based on a spectral clustering algorithm and combines eigenvector direction selection with Pearson correlation clustering in the spectral space. The method is an improvement on available spectral clustering methods, capable of robustly identifying active brain networks consistent with those from model-driven methods at different noise levels, even at the noise level of real fMRI data.

A Hierarchical Neural Network for Point Cloud Segmentation and Geometric Primitive Fitting.

Automated generation of geometric models from point cloud data holds significant importance in the field of computer vision and has expansive applications, such as shape modeling and object recognition. However, prevalent methods exhibit accuracy issues. In this study, we introduce a novel hierarchical neural network that utilizes recursive PointConv operations on nested subdivisions of point sets. This network effectively extracts features, segments point clouds, and accurately identifies and computes parameters of regular geometric primitives with notable resilience to noise. On fine-grained primitive detection, our approach outperforms Supervised Primitive Fitting Network (SPFN) by 18.5% and Cascaded Primitive Fitting Network (CPFN) by 11.2%. Additionally, our approach consistently maintains low absolute errors in parameter prediction across varying noise levels in the point cloud data. Our experiments validate the robustness of our proposed method and establish its superiority relative to other methodologies in the extant literature.

Robust baseline correction for Raman spectra by constrained Gaussian radial basis function fitting

Accurate baseline correction is a fundamental requirement for extracting meaningful spectral information and enabling precise quantitative analysis using Raman spectroscopy. Although numerous baseline correction techniques have been developed, they often require meticulous parameter adjustments and yield inconsistent results. To address these challenges, we have introduced a novel approach, namely constrained Gaussian radial basis function fitting (CGF). Our method involves solving a curve-fitting problem using Gaussian radial basis functions under specific constraints. To ensure stability and efficiency, we developed a linear programming algorithm for the proposed approach. We evaluated the performance of CGF using simulated Raman spectra and demonstrated its robustness across various scenarios, including changes in data length and noise levels. In contrast to standard methods, which frequently require complicated parameter adjustments and may exhibit varying errors, our approach provides a simple parameter search and consistently achieves low errors. We further assessed CGF using real Raman spectra, leading to enhanced accuracy in the quantitative analysis of the Raman spectra of chemical warfare agents. Our results emphasize the potential of CGF as a valuable tool for Raman spectroscopy data analysis, significantly advancing sophisticated analytical techniques.

An Assessment Of Event-Based Imaging Velocimetry For Dimensionality Reduction In Turbulent Flows

"This study investigates the feasibility of using neuromorphic event-based vision (EBV) camera for low-order modelling, a key enabler for real-time flow control. A synchronized experiment with simultaneous Event Based Image Velocimetry (EBIV) and Particle Image Velocimetry (PIV) is performed on a submerged water jet flow at Re = 2600. Flow statistics, spectral content, and reduced-order modeling capabilities using Proper Orthogonal Decomposition (POD) are assessed. Velocity fields are computed via standard PIV processing and compared after interpolation onto a common grid. The analysis reveals good agreement in jet flow statistics and spectra between EBIV and conventional PIV, with differences observed in the power spectral density (PSD) for high frequencies (St>1.5) due to higher noise levels in EBIV data. Nonetheless, most of the spectral content is correctly captured by the EBIV. EBIV successfully identifies dominant flow structures and spectral distribution of energy. The correct identification of temporal modes confirms EBIV's potential for flow control applications based on reduced order models. Despite minor discrepancies downstream of the jet in the Low Order Reconstruction (LOR) analysis, attributed to EBIV's challenge in accurately reconstructing broad-spectrum turbulent regions, the technology proves promising for real-time imaging-based flow control."

Did Short‐Term Preseismic Crustal Deformation Precede the 2011 Great Tohoku‐Oki Earthquake? An Examination of Stacked Tilt Records

AbstractThe detection of preslip, occurring hours to days before a large earthquake, using geodetic measurements has been a major focus in earthquake prediction research. A recent study claims to have detected a preseismic signal interpreted as accelerating slip near the hypocenter of the 2011 great Tohoku‐oki earthquake, starting approximately 2 hr before the mainshock. This claim is based on a stacking procedure using GNSS (Global Navigation Satellite System) data. However, a follow‐up study demonstrated that the signal disappeared when specific GNSS noise was corrected. Here we utilize tiltmeter records, independent on GNSS, to check whether the claimed preseismic signal is detected using a similar stacking procedure. Our results show no acceleration‐like deformation from 2 hr before the mainshock. This indicates that no precursory slip exceeded the noise level of the tilt data, and if any preslip occurred, it was less than 5.0 × 1018 Nm in seismic moment.

HyperGCN: an effective deep representation learning framework for the integrative analysis of spatial transcriptomics data

BackgroundAdvances of spatial transcriptomics technologies enabled simultaneously profiling gene expression and spatial locations of cells from the same tissue. Computational tools and approaches for integration of transcriptomics data and spatial context information are urgently needed to comprehensively explore the underlying structure patterns. In this manuscript, we propose HyperGCN for the integrative analysis of gene expression and spatial information profiled from the same tissue. HyperGCN enables data visualization and clustering, and facilitates downstream analysis, including domain segmentation, the characterization of marker genes for the specific domain structure and GO enrichment analysis.ResultsExtensive experiments are implemented on four real datasets from different tissues (including human dorsolateral prefrontal cortex, human positive breast tumors, mouse brain, mouse olfactory bulb tissue and Zabrafish melanoma) and technologies (including 10X visium, osmFISH, seqFISH+, 10X Xenium and Stereo-seq) with different spatial resolutions. The results show that HyperGCN achieves superior clustering performance and produces good domain segmentation effects while identifies biologically meaningful spatial expression patterns. This study provides a flexible framework to analyze spatial transcriptomics data with high geometric complexity.ConclusionsHyperGCN is an unsupervised method based on hypergraph induced graph convolutional network, where it assumes that there existed disjoint tissues with high geometric complexity, and models the semantic relationship of cells through hypergraph, which better tackles the high-order interactions of cells and levels of noise in spatial transcriptomics data.

A Generalized Formulation for Group Selection via ADMM

This paper studies a statistical learning model where the model coefficients have a pre-determined non-overlapping group sparsity structure. We consider a combination of a loss function and a regularizer to recover the desired group sparsity patterns, which can embrace many existing works. We analyze directional stationary solutions of the proposed formulation, obtaining a sufficient condition for a directional stationary solution to achieve optimality and establishing a bound of the distance from the solution to a reference point. We develop an efficient algorithm that adopts an alternating direction method of multiplier (ADMM), showing that the iterates converge to a directional stationary solution under certain conditions. In the numerical experiment, we implement the algorithm for generalized linear models with convex and nonconvex group regularizers to evaluate the model performance on various data types, noise levels, and sparsity settings.

A feasibility study of Boree Salt body mapping in the Adavale Basin using passive seismic data

Hydrogen plays a pivotal role in the global energy transition and may require underground storage. So far salt cavern storage is the only proven technology for underground hydrogen storage. The Boree Salt in the Adavale Basin, mostly at depths from 1 to 2.5 km and up to 550 m thick, consists predominantly of halite and is deemed suitable for hydrogen storage. However, current maps are inadequate. Recently passive seismic data (ambient noise) have received much interest for subsurface imaging. The main signal from passive data is surface waves (usually below 2 Hz). The capability of surface waves for the Boree Salt body mapping is examined. Parameters of seismic sensor spacing, the dominant frequencies of the surface waves, and data noise levels are all considered. It is demonstrated that surface waves from ambient noise can map the Boree Salt bodies with a survey distance of ~40 km. Between frequencies of 0.12 and 0.25 Hz, results from the latter have better resolution because of a shorter wavelength. Moving to higher frequencies of 0.5 and 1 Hz, however, the resolution becomes worse, because the depth sensitivity of surface waves moves to the shallower part of the model with increasing frequencies, rendering them incapable of effectively probing the targeted depths. For signal/noise ratio above five, station spacing can be as large as 1 km without compromising quality. Therefore, cost-effective and environmentally friendly passive seismic data can be a good alternative to the traditional active-source data for deep salt body imaging.

Assimilation of Geophysics-Derived Spatial Data for Model Calibration in Geologic CO2 Sequestration

Summary Uncertainty in geological models usually leads to large uncertainty in the predictions of risk-related system properties and/or risk metrics (e.g., CO2 plumes and CO2/brine leakage rates) at a geologic CO2 storage site. Different types of data (e.g., point measurements from monitoring wells and spatial data from 4D seismic surveys) can be leveraged or assimilated to reduce the risk predictions. In this work, we develop a novel framework for spatial data assimilation and risk forecasting. Under the U.S. Department of Energy’s National Risk Assessment Partnership (NRAP), we have developed a framework using an ensemble-based data assimilation approach for spatial data assimilation and forecasting. In particular, we took CO2 saturation maps interpreted from 4D seismic surveys as inputs for spatial data assimilation. Three seismic surveys at Years 1, 3, and 5 were considered in this study. Accordingly, three saturation maps were generated for data assimilation. The impact from the level of data noise was also investigated in this work. Our results show increased similarity between the updated reservoir models and the “ground-truth” model with the increased number of seismic surveys. Predictive accuracy in CO2 saturation plume increases with the increased number of seismic surveys as well. We also observed that with the increase in the level of data noise from 1% to 10%, the difference between the updated models and the ground truth does not increase significantly. Similar observations were made for the prediction of CO2 plume distribution at the end of the CO2 injection period by increasing the data noise.

Identifying failure types in cyber-physical water distribution networks using machine learning models

Abstract Water cyber-physical systems (CPSs) have experienced anomalies from cyber-physical attacks as well as conventional physical and operational failures (e.g., pipe leaks/bursts). In this regard, rapidly distinguishing and identifying a facing failure event from other possible failure events is necessary to take rapid emergency and recovery actions and, in turn, strengthen system's resilience. This paper investigated the performance of machine learning classification models – support vector machine (SVM), random forest (RF), and artificial neural networks (ANNs) – to differentiate and identify failure events that can occur in a water distribution network (WDN). Datasets for model features related to tank water levels, nodal pressure, and water flow of pumps and valves were produced using hydraulic model simulation (WNTR and epanetCPA tools) for C-Town WDN under pipe leaks/bursts, cyber-attacks, and physical attacks. The evaluation of accuracy, precision, recall, and F1-score for the three models in failure type identification showed the variation of their performances depending on the specific failure types and data noise levels. Based on the findings, this study discussed insights into building a framework consisting of multiple classification models, rather than relying on a single best-performing model, for the reliable classification and identification of failure types in WDNs.

Multi-source AdaBoost with cross-weight method for virtual metrology in semiconductor manufacturing

In the context of Industry 4.0, many production scenarios utilise sensors to monitor manufacturing processes, resulting in massive data that can be leveraged to build machine-learning models to improve production efficiency and quality. In semiconductor manufacturing, where many sensors are employed to monitor the production process, the resulting multi-sensor data poses challenges for constructing a virtual metrology (VM) model to assess wafer quality. Furthermore, building separate prediction models for each sensor can result in a loss of redundancy present in the multi-sensor data. Therefore, this paper proposes a multi-source AdaBoost with a cross-weight sampling method for predicting the critical dimension required in VM. Compared with the existing AdaBoost algorithm in regression, the proposed method adapts well to multi-source data by incorporating a cross-weight sampling distribution. Thus, when updating the sampling weight distribution, the impact of both individual and multi-source data is considered to re-sample high-quality observations. Based on the numerical results from the synthetic datasets and case study in semiconductor manufacturing, the proposed method outperforms benchmark methods. Additionally, owing to the redundancy of multi-sensor data, the proposed method demonstrates robust performance regardless of the noise level in the semiconductor VM data.

Impact of uncertainty in the physics-informed neural network on pressure prediction for water hammer in pressurized pipelines

In pressurized pipeline systems, accurate prediction of water hammer pressure is crucial for ensuring safe system operation. When the boundary conditions are unknown and measured data is sparse, both traditional methods fully based on physical equations and data-driven neural network methods have difficulty in accurately predicting water hammer pressure. The physics-informed neural network (PINN) overcomes these challenges by simultaneously incorporating measured data and physical equations during the network training process. However, PINN has uncertainties and their impact on the accuracy of pressure prediction is not yet clear. In this study, the valve closing water hammer in a reservoir-pipeline-valve system is taken as the research object, we investigate the influence of the uncertainty of physics and data in the PINN on prediction accuracy by using water hammer equations with various friction models and training data with various noise levels. The results show that using the water hammer equation with the Brunone model, the PINN model has higher prediction accuracy. Furthermore, data noise levels less than 10% have a relatively small impact on pressure prediction accuracy, indicating that the PINN model has good robustness in terms of data noise levels.

Deeply integrating latent consistent representations in high-noise multi-omics data for cancer subtyping.

Cancer is a complex and high-mortality disease regulated by multiple factors. Accurate cancer subtyping is crucial for formulating personalized treatment plans and improving patient survival rates. The underlying mechanisms that drive cancer progression can be comprehensively understood by analyzing multi-omics data. However, the high noise levels in omics data often pose challenges in capturing consistent representations and adequately integrating their information. This paper proposed a novel variational autoencoder-based deep learning model, named Deeply Integrating Latent Consistent Representations (DILCR). Firstly, multiple independent variational autoencoders and contrastive loss functions were designed to separate noise from omics data and capture latent consistent representations. Subsequently, an Attention Deep Integration Network was proposed to integrate consistent representations across different omics levels effectively. Additionally, we introduced the Improved Deep Embedded Clustering algorithm to make integrated variable clustering friendly. The effectiveness of DILCR was evaluated using 10 typical cancer datasets from The Cancer Genome Atlas and compared with 14 state-of-the-art integration methods. The results demonstrated that DILCR effectively captures the consistent representations in omics data and outperforms other integration methods in cancer subtyping. In the Kidney Renal Clear Cell Carcinoma case study, cancer subtypes were identified by DILCR with significant biological significance and interpretability.

Performance Analysis of Artificial Intelligence Approaches for LEMP Classification

Lightning Electromagnetic Pulses, or LEMPs, propagate in the Earth–ionosphere waveguide and can be detected remotely by ground-based lightning electric field sensors. LEMPs produced by different types of lightning processes have different signatures. A single thunderstorm can produce thousands of LEMPs, which makes their classification virtually impossible to carry out manually. The lightning classification is important to distinguish the types of thunderstorms and to know their severity. Lightning type is also related to aerosol concentration and can reveal wildfires. Artificial Intelligence (AI) is a good approach to recognizing patterns and dealing with huge datasets. AI is the general denomination for different Machine Learning Algorithms (MLAs) including deep learning and others. The constant improvements in the AI field show us that most of the Lightning Location Systems (LLS) will soon incorporate those techniques to improve their performance in the lightning-type classification task. In this study, we assess the performance of different MLAs, including a SVM (Support Vector Machine), MLP (Multi-Layer Perceptron), FCN (Fully Convolutional Network), and Residual Neural Network (ResNet) in the task of LEMP classification. We also address different aspects of the dataset that can interfere with the classification problem, including data balance, noise level, and LEMP recorded length.

Assessing the impact of turbulent kinetic energy boundary conditions on turbulent flow simulations using computational fluid dynamics

Computational fluid dynamics has been widely used to study hemodynamics, but accurately determining boundary conditions for turbulent blood flow remains challenging. This study aims to investigate the effect of patient-specific turbulence boundary conditions on the accuracy of turbulent flow simulation. Using a stenosis model with 50% severity in diameter, the post-stenosis turbulence flow region was simulated with different planes to obtain inlet boundary conditions and simulate downstream flows. The errors of simulated flow fields obtained with turbulence kinetic energy (TKE) boundary data and arbitrary turbulence intensity were compared. Additionally, the study tested various TKE data resolutions and noise levels to simulate experimental environments. The mean absolute error of velocity and TKE was investigated with various turbulence intensities and TKE mapping. While voxel size and signal-to-noise ratio of the TKE data affected the results, simulation with SNR > 5 and voxel size < 10% resulted in better accuracy than simulations with turbulence intensities. The simulation with appropriate TKE boundary data resulted in a more accurate velocity and turbulence field than those with arbitrary turbulence intensity boundary conditions. The study demonstrated the potential improvement of turbulent blood flow simulation with patient-specific turbulence boundary conditions, which can be obtained from recent measurement techniques.

ScTIGER: A Deep-Learning Method for Inferring Gene Regulatory Networks from Case versus Control scRNA-seq Datasets.

Inferring gene regulatory networks (GRNs) from single-cell RNA-seq (scRNA-seq) data is an important computational question to find regulatory mechanisms involved in fundamental cellular processes. Although many computational methods have been designed to predict GRNs from scRNA-seq data, they usually have high false positive rates and none infer GRNs by directly using the paired datasets of case-versus-control experiments. Here we present a novel deep-learning-based method, named scTIGER, for GRN detection by using the co-differential relationships of gene expression profiles in paired scRNA-seq datasets. scTIGER employs cell-type-based pseudotiming, an attention-based convolutional neural network method and permutation-based significance testing for inferring GRNs among gene modules. As state-of-the-art applications, we first applied scTIGER to scRNA-seq datasets of prostate cancer cells, and successfully identified the dynamic regulatory networks of AR, ERG, PTEN and ATF3 for same-cell type between prostatic cancerous and normal conditions, and two-cell types within the prostatic cancerous environment. We then applied scTIGER to scRNA-seq data from neurons with and without fear memory and detected specific regulatory networks for BDNF, CREB1 and MAPK4. Additionally, scTIGER demonstrates robustness against high levels of dropout noise in scRNA-seq data.

Noise-robust estimation of the maximal Lyapunov exponent based on state space reconstruction with principal components

The maximal Lyapunov exponent (MLE) is a widely used indicator of the dependence on initial conditions of various systems in many fields. However, both the accuracy and the consistency of the conventional method for estimating MLE heavily depend on the noise level. We developed a method for estimating MLE with higher accuracy and noise-robustness compared with the conventional method. Considering that state space reconstruction using principal components can enhance noise-robustness when appropriate parameter values are used, we devised a set of algorithms for finding proper window length and the number of principal components required for state space reconstruction and MLE estimation. Numerical simulations of multiple dynamical systems with various data lengths and noise levels verified that the devised algorithm yields more accurate, consistent, and noise-robust estimation of MLE than the conventional method.

Retrieval of parameters in micromixer with obstacle by cascade‐forward‐type artificial neural network: Comparison of four training algorithms under noisy data

AbstractTwo parameters are retrieved in a passive Y‐type micromixer with circular obstacle by cascade‐forward‐type artificial neural network (CFANN). The governing equations are solved by the finite volume method, under specific boundary conditions. The numerical model is then used to compute velocity profile and mixing efficiency, for different values of the Reynolds number. Thus, the velocity profiles along with Reynolds number (Re) and mixing efficiency (η) constitute the input–output pair of data. These data are used to train CFANN, and the network is monitored through different means, like, histograms, performance curves, and so forth. For inverse analysis, the trained CFANN model is fed with a new velocity profile as input, and corresponding values of Reynolds number and mixing efficiency are obtained as output. In an attempt to construct the optimum CFANN model, various combinations were explored, like, (1) different numbers of neurons in the hidden layer, (2) different noise levels in input data, and (3) different algorithms in the training stage. Finally, the CFANN with 10 hidden layer neurons with Levenberg–Marquardt (LM) algorithm was found to give retrieved values with up to 0.96% absolute error for all levels of noise in the input data. Also, the CFANN model with the LM algorithm has a very high value of regression coefficient of greater than 0.998, under all the noise values. Scaled conjugate gradient algorithm gives good results for the no‐noise case, but fails poorly with the rise of noise. Other algorithms, like, Bayesian regularization and resilient backpropagation, perform poorly even in the no‐noise case. The present approach is highly simple, accurate, and time efficient for applying inverse analysis in micromixers.

전력시스템 진동 해석을 위한 RMS 필터링 기반 Prony 분석 차수 선택

This paper aims to investigate RMS filtering and Prony analysis for detecting oscillation frequencies in order to evaluate the stability and reliability of power systems, proposing a method for selecting the polynomial order in Prony analysis. Among the oscillation detection methods using data, Prony analysis is commonly utilized due to its precision and capability of providing abundant information such as oscillation frequency and damping degree, but there is no standard for selecting the polynomial order. The order needs to be selected based on factors such as data volume, sampling interval, and noise level. In this paper, the Hann Window is applied in each section to select the appropriate order and the dominant oscillation mode is estimated through RMS filtering. The order that derives the estimated frequency results is applied in Prony analysis. The proposed method is demonstrated to be effective with known test data and actual measurement data.