Main Dataset Research Articles

Icelandic translation, adaptation and validation of the Parental Burnout Assessment (PBA-IS)

Background As parental burnout is increasingly recognised for its severe impact on parents and children, identifying factors that exacerbate or alleviate this condition is crucial. Reliable assessment tools in clinical settings are essential to detect those at risk of or experiencing burnout, enabling timely intervention. Aims/objectives This study aims to adapt the Parental Burnout Assessment for use in Iceland and evaluate its psychometric properties while exploring how personal and socio-demographic factors influence parental burnout. Materials and methods A sample of 1,110 parents participated. Descriptive statistics analysed the main dataset characteristics, and confirmatory factor analysis evaluated the psychometric properties of the adapted version. Results Satisfactory structural validity and internal consistency (α 0.96) of the PBA-IS was demonstrated. Factors influencing parental burnout included marital status, number of children, perceived support, and personal causation. Conclusions The PBA-IS is a valid and reliable translated tool for assessing parental burnout in Iceland. Personal causation, a key concept in occupational therapy, appears pivotal in parental burnout. Occupational therapists can provide holistic support to help parents effectively manage stress. Significance The PBA-IS enables parental burnout to be identified in Icelandic clinical settings, supporting early interventions that reduce stress, promote mental health, and enhance well-being.

Adopting random forest for predicting the risk of cerebrovascular disease and diabetes using appropriate database

Random forest is used to predict the possibility of the existence of a cerebrovascular disease subjects curated from the BraVa and diabetes datasets. Towards analyzing and prediction of cerebrovascular diseases, SPSS for realizing the independence and correlation between the various metrics that could contribute to a subject been diagnosed with a cerebrovascular disease. An analysis of the various metrics of the overall vascular size revealed a significant correlation especially between Total Length and Total Number of Branches (R = 0.829, p = 0.000). Metrics like Age, Contraction, Tortuosity, mean bifurcation Angle, mean bifurcation tilt which has implication of a cerebrovascular disease diagnosis according to study was used as the input for the random forest algorithm. The BraVa dataset which is the main datasets for this work was used to train the algorithm and a prediction of either “risky” or “Not risky” with a high accuracy of 100% was recorded. To further test the algorithm, a second datasets from the from the diabetes database which has a high number of subjects was also used to test the algorithm and a high accuracy of 90.256% was recorded. It was determined from the results that machine learning based Random Forest algorithm can be adopted as a prediction method especially on bigger dataset of neuromorphological measurements of neurons and it will aid or facilitate accurate prediction of any form of cerebrovascular disease and also aid in accurate medical diagnosis.

QM40, Realistic Quantum Mechanical Dataset for Machine Learning in Molecular Science

The growing popularity of machine learning (ML) and deep learning (DL) in scientific fields is hindered by the scarcity of high-quality datasets. While quantum mechanical (QM) predictions using DL techniques such as graph neural networks (GNNs) and generative models are gaining traction, insufficient training data remains a bottleneck. The QM40 dataset addresses this challenge by representing 88% of the FDA-approved drug chemical space. It includes molecules containing 10 to 40 atoms and composed of elements commonly found in drug molecular structures (C, O, N, S, F, Cl). QM40 offers valuable resources for researchers which include the core QM40 main dataset, containing 16 key quantum mechanical parameters for 162,954 molecules calculated using the B3LYP/6-31G(2df,p) level of theory in Gaussian16, ensuring consistency with established datasets like QM9 and Alchemy. This compatibility allows for future concatenation of QM40 with these datasets. In addition to other valuable information, the QM40 dataset offers the initial and optimized Cartesian coordinates, Mulliken charges, and detailed bond information, including local vibrational mode force constants, which serve as indicators of bond strength. QM40 can be used to benchmark both existing and new methods for predicting QM calculations using ML and DL techniques.

CRNet: Cascaded Refinement Network for polyp segmentation

Technology for automatic segmentation plays a crucial role in the early diagnosis and treatment of ColoRectal Cancer (CRC). Existing polyp segmentation methods often focus on advanced feature extraction while neglecting detailed low-level features, This somewhat limits the enhancement of segmentation performance. This paper proposes a new technique called the Cascaded Refinement Network (CRNet), designed to improve polyp segmentation performance by combining low-level and high-level features through a cascaded contextual network structure. To accurately capture the morphological variations of polyps and enhance the clarity of segmentation boundaries, we have designed the Multi-Scale Feature Optimization (MFO) module and the Contextual Edge Guidance (CEG) module. Additionally, to further enhance feature fusion and utilization, we introduced the Cascaded Local Feature Fusion (CLFF) module, which effectively integrates cross-layer correlations, allowing the network to understand complex polyp structures better. By conducting a large number of experiments, our model achieved a 0.3% and 3.1% higher mDice score than the latest MMFIL-Net in the two main datasets of Kvasir-SEG and CVC-ClinicDB, respectively. Ablation studies show that MFO improves the baseline score by 4%, and the network without CLFF and CEG results in a reduction of 2.4% and 1.7% in mDice scores, respectively. This further validates the contribution of each module to the polyp segmentation performance. CRNet enhances model performance through the introduction of multiple modules but also increases model complexity. Future work will explore how to reduce computational complexity and improve inference speed while maintaining high performance. The source code for this paper can be found at https://github.com/l1986036/CRNet.

Improving Rainfall Prediction Accuracy in the USA Using Advanced Machine Learning Techniques

The key aim of this research project is to design and evaluate advanced machine learning models for increasing accuracy in rainfall forecasting over the USA. We intended to investigate nonlinear relationships typical of the atmospheric variables using state-of-the-art ML methods for more accurate rainfall predictions. For this research project on rainfall forecasting in the USA, we utilized an extensive dataset that comprises historical rainfall data collected from the National Oceanic and Atmospheric Administration (NOAA) and other meteorological agencies. The main dataset we use in this paper consists of daily rainfall measurements across various geographical locations of the USA, thus capturing the wide-ranging historical data necessary for both training and validation of the model. Besides measuring rainfall, we included meteorological data from sources such as NOAA's Global Historical Climatology Network and NASA's Modern-Era Retrospective Analysis for Research and Applications. These datasets further provided key variables that are known to affect rain, including temperature, humidity, wind speed, and atmospheric pressure. The performance metrics used in this research work for the models considered include accuracy, precision, recall, and F1 score. The above table shows that the Random Forest Classifier outperformed the other models, achieving perfect accuracy. That indicated that it rightly classified all the instances in the test set. The Logistic Regression and Support Vector Machine models gave a quite good performance by giving above average accuracy but had lower precision and recall for the rainfall prediction. Accurate rainfall forecasting has direct consequences on agriculture, greatly empowering farmers and agricultural planners to make more effective decisions regarding planting, harvesting, and crop management. The forecasts of rainfall are also of critical importance in disaster management regarding planning for flood emergencies. Moreover, precise forecasting of rainfall, particularly in sustainable water resources management, presents the most important data in planning for and conserving these resources.

COVID-19 pandemic microplastics environmental impacts predicted by deep random forest (DRF) predictive model

BackgroundMicroplastic pollution is a pressing issue with far-reaching environmental and public health consequences. This study delves into the intricacies of predicting microplastic pollution during the COVID-19 pandemic in Tehran, Iran.MethodsThe research introduces a rigorous comparative analysis that evaluates the predictive prowess of the Deep Random Forest algorithm and established benchmarks, such as Random Forest, Decision Trees, Gradient Boosting, AdaBoost, and Support Vector Machine. The evaluation process encompasses a meticulous 70–30 training–testing split of the main data set. Performance is assessed by analysis metrics, including ROC and statistical errors. The primary data set encompasses distinct categories, including household wastes, hospital wastes, clinics wastes, and unknown-originated susceptible waste which is categorized in Infected items, PPEs, SUPs, Test kits, Medical packages, Unknown-originated pandemic mircoplastic waste. Deliberately, this data set was partitioned into training and testing subsets, ensuring the robustness and reliability of subsequent analyses. Approximately 70% of the main database was allocated to the training data set, with the remaining 30% constituting the testing data set.ResultsThe findings underscore the proposed algorithm’s supremacy, boasting an impressive AUC = 0.941. This exceptional score reflects the model’s precision in categorizing microplastics. These results have profound implications for environmental management and public health during pandemics.ConclusionsThe study positions the proposed model as a potent tool for microplastic pollution prediction, encouraging further research to refine predictive models and tap into new data sources for a more comprehensive understanding of microplastic dynamics in urban settings.

Beyond the hashtags: social media usage and cryptocurrency investment

PurposeCryptocurrencies have gained popularity among investors despite their high risk and volatility. Social media wields substantial influence over investors' attitudes, judgments and decisions related to investment. This study aims to investigate the associations between social media usage and cryptocurrency investment behavior.Design/methodology/approachUtilizing the main dataset of the 2021 National Financial Capability Study and its supplementary Investor Survey, this study analyzed social media usage in general. Additionally, it separately examined 11 different social media platforms as potential sources of information for investments. Logistic regressions were performed to explore the relationship between social media, previous experiences and future considerations in investing in cryptocurrencies. Robustness checks were conducted with additional analyses.FindingsInvestors who used social media for investment information were more likely to invest in cryptocurrencies and consider investing in cryptocurrencies in the future. The likelihood increased with the number of social media platforms used. Different social media platforms exhibited distinct associations with cryptocurrency investment experiences and future considerations.Originality/valueThis study is one of the initial attempts to examine the role of social media platforms in cryptocurrency investment. The findings offer unique and important theoretical and practical insights for policymakers, researchers and practitioners, which can benefit consumer well-being.

Change of Ownership and Quality of Home Health Agency Care

The home health agency (HHA) market has seen increasingly more change in ownership transactions. Little is known about the organizational characteristics and quality of care of HHAs after ownership has changed. To examine whether an HHA change in ownership was associated with different quality-of-care outcomes, patient volume, and staffing levels. Using current HHA change of ownership files linked to publicly available Medicare HHA data, this staggered difference-in-differences analysis evaluated ownership change transactions of Medicare-certified HHAs from quarter 1 of 2016 to quarter 4 of 2019. Pretransaction and posttransaction HHA characteristics and quality-of-care outcomes were compared between HHAs that changed ownership and up to 8 matched controls. The transactions in the main analysis included (1) when an HHA remained as for-profit with a change in ownership and (2) when an HHA changed from nonprofit/public to for-profit ownership. The data were analyzed between November 2023 and September 2024. The primary outcomes were HHA-quarter measures of star ratings, the individual quality measures that compose the star ratings based on the Outcome and Assessment Information Set, and claims-based quality metrics, such as hospitalizations and emergency department visits. The secondary outcomes included HHA year measures of Medicare patient volume, per capita payments, and staffing levels (full-time equivalents and minutes per visit). The main dataset included 294 Medicare-certified HHAs that changed ownership, matched with 2330 controls. In 3 years after an ownership change, quarterly star ratings increased by 0.18 (95% CI, 0.05-0.31) relative to matched controls, with greater increases among HHAs that converted from nonprofit/public to for-profit. No significant improvement was observed in the 60-day rates of hospital admissions or outpatient emergency department visits. Further, no significant changes were observed in the number of Medicare beneficiaries, but per capita payments increased within 2 years post-ownership change. Significant reductions were observed in full-time equivalents of registered nurses (-17% [95% CI, -31% to -3%]) and home health aides (-26% [95% CI, -39% to -13%]), as well as per-visit minutes for skilled nursing care (-5% [95% CI, -9% to -1%]), physical therapy (-3% [95% CI, -5% to 0%]), and home health aide care (-11% [95% CI, -15% to -6%]). In this difference-in-differences analysis of Medicare-certified HHAs, ownership change was associated with higher star ratings and Medicare per capita payments, but not with claims-based quality measures. Reduction in staffing levels after ownership change raises concerns about implications for quality of care.

A Novel Approach to Enhancing the Accuracy of Prediction in Ship Fuel Consumption

Ship fuel consumption plays a crucial role not only in understanding ships’ energy efficiency but also in gaining insights into their emissions. However, enhancing the accuracy of these predictions poses significant challenges due to data limitations and the methods employed. Due to these factors, such as data variability and equipment characteristics, ship fuel consumption exhibits certain fluctuations under specific conditions. Previous fuel consumption prediction methods primarily generate a single specific value, making it difficult to capture the volatility of and variability in fuel consumption. To overcome this limitation, this paper proposes a novel method that integrates Gaussian process prediction with quantile regression theory to perform interval predictions of ship fuel consumption, providing a range of possible outcomes. Through comparative analyses with traditional methods, the possibility of using the method is verified and its results are validated. The results indicate the following: (1) at a 95% confidence level, the proposed method achieves a prediction interval coverage probability of 0.98 and a prediction interval normalized average width of 0.123, which are significantly better than those of the existing backpropagation neural network (BPNN) and gradient boosting decision tree (GBDT) quantile regression models; (2) the prediction accuracy of the proposed method is 92% for point forecasts; and (3) the proposed method is applicable to main datasets, including both noon report and sensor datasets. These findings provide valuable insights into interval predictions of ship fuel consumption and highlight their potential applications in related fields, emphasizing the importance of accurate interval predictions in intelligent energy efficiency optimization.

Methods to Handle Missing depression (PHQ-8 Score) Values among Adults in the United Arab Emirates

Abstract Background The UAE Healthy Future Study (UAEHFS) is one of the first large prospective cohort studies in the Gulf region which examines causes and risk factors for chronic diseases among adult UAE nationals. Missing values are often unavoidable in empirical research and can in many cases, lead to bias. The aim of this study is to estimate the percentage of depression in the UAEHFS pilot data using the eight-item Patient Health Questionnaire (PHQ-8) variables, using different statistical methods. Methods Five common statistical machine learning methods of handling missing values were included in this analysis. These are mode imputation, k-nearest neighbor (KNN) imputation, classification, and regression trees (CART), random forest (RF) imputations, and random sample from observed values (Sample). 100 multiple imputations were used. Results 487 (94.2 %) eligible participants were included in the analysis. 231 (44.7%) were included in the complete case analysis. The median age was 30 years (Interquartile-Range: 23 - 38). More males (67.8%) than females included in the analysis. The estimated percentage of depression was 8.4%, 8.9%, 9.9%, 12.5%, 15.4% and 17.9% by the mode, complete case, sample, RF, CART, and KNN respectively. In additional analyses, the estimated proportions of depression were 11.5% by the Complete Case, 11.9% by KNN, 13.2% by K-means clustering, and 13.2% by Random Forest. Conclusions The estimated percentage of depression in the UAEHFS pilot data varies between the applied methods of handling missing values. This shows that the problem of missing values in the variables is not negligible. Further research is needed using multiple imputations in the main UAEHFS dataset after completing recruitment. Key messages • For the depression missing values, we recommend using multiple imputations not to generate data but to prevent the exclusion of observed data. • To have a better estimate of the percentage of depression, is recommended to use different machine learning methods.

Temperature effect on seawater fCO2 revisited: theoretical basis, uncertainty analysis and implications for parameterising carbonic acid equilibrium constants

Abstract. The sensitivity of the fugacity of carbon dioxide in seawater (fCO2) to temperature (denoted υ, reported in % °C−1) is critical for the accurate fCO2 measurements needed to build global carbon budgets and for understanding the drivers of air–sea CO2 flux variability across the ocean. However, understanding and computing υ have been restricted to either using empirical functions fitted to experimental data or determining it as an emergent property of a fully resolved marine carbonate system, and these two approaches are not consistent with each other. The lack of a theoretical basis and an uncertainty estimate for υ has hindered resolving this discrepancy. Here, we develop a new approach for calculating the temperature sensitivity of fCO2 based on the equations governing the marine carbonate system and the van 't Hoff equation. This shows that, to first order, ln (fCO2) should be proportional to 1/tK (where tK is temperature in kelvin), rather than to temperature, as has previously been assumed. This new approach is, to first order, consistent with calculations from a fully resolved marine carbonate system, which we have incorporated into the PyCO2SYS software. Agreement with experimental data is less convincing but remains inconclusive due to the scarcity of direct measurements of υ, particularly above 25 °C. However, the new approach is consistent with field data, performing better than any other approach for adjusting fCO2 by up to 10 °C if spatiotemporal variability in its single fitted coefficient is accounted for. The uncertainty in υ arising from only measurement uncertainty in the main experimental dataset where υ has been directly measured is in the order of 0.04 % °C−1, which corresponds to a 0.04 % uncertainty in fCO2 adjusted by +1 °C. However, spatiotemporal variability in υ is several times greater than this, so the true uncertainty due to the temperature adjustment in fCO2 adjusted by +1 °C using the most widely used constant υ value is around 0.24 %. This can be reduced to around 0.06 % using the new approach proposed here, and this could be further reduced by more measurements. The spatiotemporal variability in υ arises mostly from the equilibrium constants for CO2 solubility and carbonic acid dissociation (K1∗ and K2∗), and its magnitude varies significantly depending on which parameterisation is used for K1∗ and K2∗. Seawater fCO2 can be measured accurately enough that additional experiments should be able to detect spatiotemporal variability in υ and distinguish between different parameterisations for K1∗ and K2∗. Because the most widely used constant υ was coincidentally measured from seawater with roughly global average υ, our results are unlikely to significantly affect global air–sea CO2 flux budgets, but they may have more important implications for regional budgets and studies that adjust by larger temperature differences.

Damage characterisation of GFRP composites based on clustering acoustic emission events utilizing single-failure-cause tests as reference

A new method to identify causes of fracture in composites based on acoustic emission (AE) and clusterization of AE data based on reference datasets is presented within the manuscript. Acoustic Emission (AE) is a widely used non-destructive method for fracture analysis, but data due to their multidimensionality are not easy to analyze especially if the acoustic events appear simultaneously and have similar parameters even if they are an effect of different failure mechanisms. In this research, we utilize an unsupervised learning algorithm besides the simplest K-means, through fuzzy c-means to Gaussian Mixture Model (GMM) and spectral clustering to investigate the dataset obtained from the three-point bending test manufactured by us composite. The analysis is preceded by data curation, feature determination (Laplacian score) and the best number of cluster investigations (DB index, Caliński-Harabasz score, and Silhouette method) To enable interpretation of the clustering we run an additional three groups of tests covering fibre breakage (two methods), resin fracture (in tension and in compression) and delamination (DCB test) creating reference datasets. These datasets were statistically analyzed and kernel density estimators were generated for each AE feature as well as amplitude-frequency characteristics. Clusters obtained for the main dataset were then assigned to particular causes of failure by comparing them with the reference dataset. It was found that clusters generated using spectral clustering were the most realistic ones, as it was possible to assign the cause of failure to them.

Quantum Deep Learning: A Review

Quantum computing is described as a process by which a system calculates output. Quantum physics usually refers to the smallest discrete unit of any property; the basic unit of data in quantum is the qubit; this qubit unit is equivalent to the bit unit in classical neural networks. Quantum deep learning combines quantum computing with deep learning to reduce training time for neural networks, which has proven effective in solving some intractable problems on classical computers. Quantum deep learning has proven effective in solving some intractable problems on classical computers. A quantum network can benefit from quantum informationflow because it is a more efficient framework than classical systems. Each quantum deep learning consists of a quantum gate. In this review, we provide a comprehensive review of recent studies that include different quantum deep learning applications, including healthcare, handwriting, and many others. Also, methodologies, problems, main datasets, results, strengths, limitations, and challenges are included in this review.

Probabilistic seismic hazard analysis for the Dominican Republic

The Dominican Republic experiences moderate to high seismic hazard mostly caused by oblique convergence at the Caribbean/North American plate boundary that manifests as subduction zones, less-pronounced subduction-like trenches with thrust faulting, long strike-slip faults parallel to the plate boundary, and onshore deformation. Historical earthquakes have damaged the Dominican Republic’s large cities and those in neighboring Haiti, once requiring relocation. Given this, the Dominican Republic joined the “Training and Communication for Earthquake Risk Assessment” (TREQ) project funded by the United States Agency for International Development, which aimed to increase earthquake risk assessment capacity in Latin American cities. The TREQ project was the basis for developing an openly available probabilistic seismic hazard model for the Dominican Republic. The input model was developed from two main datasets: a homogenized earthquake catalog and an active faults database that combines results of recent local projects with a global database. The seismic source characterization used these to constrain source geometries and occurrence rates for active shallow crustal earthquakes, subduction interfaces and subduction-like thrusts, and intraslab earthquakes. Shallow crustal earthquakes, including those on subduction-like thrusts, are modeled by smoothed seismicity and fault sources, the latter using pre-defined geometries that permit multi-fault ruptures. Seismicity on the Puerto Rico Trench subduction interface is modeled as a fault source, while intraslab sources use pre-defined gridded ruptures inside the intraslab volume. The source characterization applies epistemic uncertainties to modeling assumptions affecting occurrence rates and maximum magnitudes. The ground motion characterization used residual analyses from past regional projects as a basis, updating some components with more recent ground motion models. Computed hazard results reinforce those from recent studies in terms of geographical hazard patterns and levels. For 475-year return periods, peak ground acceleration (PGA) in Santiago de los Caballeros reaches 0.50 g, controlled by the Septentrional Fault, while all tectonic region types contribute to the PGA 0.31 g computed for Santo Domingo.

Growth of Digital Entrepreneurship in 2014~2023: A Bibliometric Analysis

Objective: Conducting a bibliometric analysis in digital entrepreneurship, dataset is from 2014 ~ 2023. Theoretical Framework: Bibliometric analysis, an analysis method within scient metrics, employs mathematical and statistical techniques to examine and assess journal articles, thereby identifying patterns and structures in particular research areas. Method: The bibliometric analysis method was conducted in this research, and the main dataset was from SCOPUS. It examined the trends, keywords, and collaborations of published articles. Results and Discussion: VOSViewer was applied to analyse 352 published articles. The results presented a significant increase in research output in field of digital entrepreneurship. Research Implications: This section presents the findings of the bibliometric analysis. It includes a detailed examination of research trends, noting a significant increase in the number of publications on digital entrepreneurship from 2019 onwards. The analysis identifies key themes and trends within the literature, such as the impact of digital technologies on business and entrepreneurship, the emergence of digital startups, and the role of digitalization in transforming SMEs. The results also highlight the contributions of various institutions to the field, indicating a broad international engagement with digital entrepreneurship research. Originality/Value: The conclusion summarizes the key findings of the study, emphasizing the growing academic interest and engagement in digital entrepreneurship. It underscores the importance of digital technologies in driving innovation and economic growth, and the need for further research to explore new opportunities and address the challenges in this evolving field.

ACP-PDAFF: Pretrained model and dual-channel attentional feature fusion for anticancer peptides prediction

Anticancer peptides(ACPs) have attracted significant interest as a novel method of treating cancer due to their ability to selectively kill cancer cells without damaging normal cells. Many artificial intelligence-based methods have demonstrated impressive performance in predicting ACPs. Nevertheless, the limitations of existing methods in feature engineering include handcrafted features driven by prior knowledge, insufficient feature extraction, and inefficient feature fusion. In this study, we propose a model based on a pretrained model, and dual-channel attentional feature fusion(DAFF), called ACP-PDAFF. Firstly, to reduce the heavy dependence on expert knowledge-based handcrafted features, binary profile features (BPF) and physicochemical properties features(PCPF) are used as inputs to the transformer model. Secondly, aimed at learning more diverse feature informations of ACPs, a pretrained model ProtBert is utilized. Thirdly, for better fusion of different feature channels, DAFF is employed. Finally, to evaluate the performance of the model, we compare it with other methods on five benchmark datasets, including ACP-Mixed-80 dataset, Main and Alternate datasets of AntiCP 2.0, LEE and Independet dataset, and ACPred-Fuse dataset. And the accuracies obtained by ACP-PDAFF are 0.86, 0.80, 0.94, 0.97 and 0.95 on five datasets, respectively, higher than existing methods by 1% to 12%. Therefore, by learning rich feature informations and effectively fusing different feature channels, ACD-PDAFF achieves outstanding performance. Our code and the datasets are available at https://github.com/wongsing/ACP-PDAFF.

Adaptive detectors for mismatched subspace target in clutter with lognormal texture

Due to factors such as array misalignment and waveform distortion, the target echo may not be precisely located within the nominal subspace in practical applications, resulting in mismatch issues. In response to this challenge, this paper investigates adaptive detection methods for detecting mismatched subspace targets amidst lognormal clutter background. To enhance the suppression of mismatched signal, we introduce a fictitious signal into the null hypothesis, which is situated within a subspace orthogonal to the nominal subspace in the whitened observation space. Following convention, we allow for the existence of a training dataset that shares the same covariance matrix (CM) structure as the main dataset. Subsequently, we propose two adaptive subspace detectors based on a two-step Generalized Likelihood Ratio Test (GLRT) and a two-step maximum a posteriori (MAP) GLRT. Both novel detectors have been validated to have constant false alarm rate (CFAR) properties for speckle CM. Numerical experiments are carried out using simulation data and measured sea clutter data, which demonstrate that our proposed methods exhibit robustness against non-mismatched signal and effective suppression of mismatches.

Optimized Multimodal Emotional Recognition Using Long Short-Term Memory

The aim of this project is to research and classification on human emotions. A new method for the recognition of speech signals has been introduced. It’s called LSTM (Long-Short Term Memory). It is a type of Recurrent neural network. RNN is used for analyzing sequential data, hence it is useful for speech signal recognition. Several Datasets were found across the internet for this project. Ex: TESS (Toronto Emotional Speech Set), RAVDESS (Ryerson Audio-Visual Database of Emotional Speech and Song), SAVEE (Surrey Audio-Visual Expressed Emotion), CREMA-D (Crowd- Sourced Emotional Multimodal Actors Dataset). The Main Dataset used in this project is TESS (Toronto Emotional Speech Set) Dataset and Mel Frequency Cepstral Coefficient (MFCC) is Used for Feature extraction.

Optimizing anomaly detection in 3D MRI scans: The role of ConvLSTM in medical image analysis

The analysis of Medical Images (MI), particularly the detection and classification of anomalies in 3D MRI (Magnetic Resonance Imaging) scans, plays a critical part in timely intervention and personalized therapy plans. In our paper, a comprehensive methodology for anomaly detection in 3D MRI scans of the brain is proposed that combines advanced Deep Learning (DL) techniques, including Convolutional Long Short-Term Memory (ConvLSTM) model, with efficient statistical feature extraction from segmented anomaly regions. The data collection phase utilizes three main datasets namely the Brain Tumor Segmentation Challenge (BRATS), the Federated Tumor Segmentation Challenge (FETS), and the Medical Segmentation Decathlon (MSD). The research begins with preprocessing steps including image resizing, intensity normalization, and alignment of anomaly region segmentation masks. The targeted anomaly regions within the samples are segmented using U-Net architecture and then follow statistical feature extraction procedure. Dimensionality reduction methods such as Principal Component Analysis (PCA) and Recursive Feature Elimination (RFE) are utilized to streamline the feature space. The ConvLSTM is then used to classify the anomalies using both convolution and recurrent layers to capture spatiotemporal patterns in MRI data. The model is fine-tuned and iterated for better classification performance using Adam optimizer. The statistical evaluation results with accuracy of 98.9 % showed that the designed ConvLSTM method is more suitable for clinical diagnosis and treatment design in detecting anomalies in 3D MRI images.

Galaxy merger challenge: A comparison study between machine learning-based detection methods

Aims. Various galaxy merger detection methods have been applied to diverse datasets. However, it is difficult to understand how they compare. Our aim is to benchmark the relative performance of merger detection methods based on machine learning (ML). Methods. We explore six leading ML methods using three main datasets. The first dataset consists of mock observations from the IllustrisTNG simulations, which acts as the training data and allows us to quantify the performance metrics of the detection methods. The second dataset consists of mock observations from the Horizon-AGN simulations, introduced to evaluate the performance of classifiers trained on different, but comparable data to those employed for training. The third dataset is composed of real observations from the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP) survey. We also compare mergers and non-mergers detected by the different methods with a subset of HSC-SSP visually identified galaxies. Results. For the simplest binary classification task (i.e. mergers vs. non-mergers), all six methods perform reasonably well in the domain of the training data. At the lowest redshift explored 0.1 < ɀ < 0.3, precision and recall generally range between ~70% and 80%, both of which decrease with increasing ɀ as expected (by ~5% for precision and ~10% for recall at the highest ɀ explored 0.76 < ɀ < 1.0). When transferred to a different domain, the precision of all classifiers is only slightly reduced, but the recall is significantly worse (by ~20–40% depending on the method). Zoobot offers the best overall performance in terms of precision and F1 score. When applied to real HSC observations, different methods agree well with visual labels of clear mergers, but can differ by more than an order of magnitude in predicting the overall fraction of major mergers. For the more challenging multi-class classification task to distinguish between pre-mergers, ongoing-mergers, and post-mergers, none of the methods in their current set-ups offer good performance, which could be partly due to the limitations in resolution and the depth of the data. In particular, ongoing-mergers and post-mergers are much more difficult to classify than pre-mergers. With the advent of better quality data (e.g. from JWST and Euclid), it is of great importance to improve our ability to detect mergers and distinguish between merger stages.