Current Dataset Research Articles

GEEES: Inferring Cell-specific Gene-Enhancer Interactions from Multi-modal Single Cell Data.

Gene-enhancer interactions are central to transcriptional regulation. Current multi-modal single cell datasets that profile transcriptome and chromatin accessibility simultaneously in a single cell are yielding opportunities to infer gene-enhancer associations in a cell type specific manner. Computational efforts for such multi-modal single cell datasets thus far focused on methods for identification and refinement of cell types and trajectory construction. While initial attempts for inferring gene-enhancer interactions have emerged, these have not been evaluated against benchmark datasets that materialized from bulk genomic experiments. Furthermore, existing approaches are limited to inferring gene-enhancer associations at the level of grouped cells as opposed to individual cells, thereby ignoring regulatory heterogeneity among the cells. We present a new approach, GEEES for "Gene EnhancEr IntEractions from Multi-modal Single Cell Data", for inferring gene-enhancer associations at the single cell level using multi-modal single cell transcriptome and chromatin accessibility data. We evaluated GEEES alongside several multivariate regression-based alternatives we devised and state-of-the-art methods using a large number of benchmark datasets, providing a comprehensive assessment of current approaches. This analysis revealed significant discrepancies between gold-standard interactions and gene-enhancer associations derived from multi-modal single-cell data. Notably, incorporating gene-enhancer distance into the analysis markedly improved performance across all methods, positioning GEEES as a leading approach in this domain. While the overall improvement in performance metrics by GEEES is modest, it provides enhanced cell representation learning which can be leveraged for more effective downstream analysis. Furthermore, our review of existing experimentally driven benchmark datasets uncovers their limited concordance, underscoring the necessity for new high-throughput experiments to validate gene-enhancer interactions inferred from single-cell data. https://github.com/keleslab/GEEES. Supplementary data are available at Bioinformatics online.

Exploring the longitudinal expression dynamics of midguts in adult female Amblyomma americanum ticks

BackgroundFemale ticks remain attached to their host for multiple days to complete a blood meal. This prolonged feeding period is accompanied by a significant increase in the tick’s size and body weight, paralleled by noteworthy changes to the tick midgut. While the midgut is recognized for its established role in blood storage and processing, its importance extends to playing a crucial role in the acquisition, survival, and proliferation of pathogens. Despite this, our overall understanding of tick midgut biology is limited.ResultsOur transcriptome analysis identified 15,599 putative DNA coding sequences (CDS), which were classified into 26 functional groups. Dimensional and differential expression analyses revealed four primary transcriptional profiles corresponding to unfed, slow-feeding, transitory (from slow- to rapid-feeding), and rapid-feeding stages. Additionally, comparing the current dataset with previously deposited transcriptome from other tick species allowed the identification of commonly expressed transcripts across different feeding stages.ConclusionOur findings provide a detailed temporal resolution of numerous metabolic pathways in the midgut of A. americanum adult females throughout the feeding process, highlighting the dynamic transcriptional regulation of the tick’s midgut as feeding progresses. Furthermore, we identified conserved transcripts across three different tick species that exhibit similar expression patterns. This knowledge not only enhances our understanding of the physiological processes within the tick midgut but also opens up potential avenues for developing control methods that target multiple tick species.

Reviewing Open Data Semantic 3D City Models to Develop Novel 3D Reconstruction Methods

Abstract. Structured semantic 3D city models are pivotal in creating urban 3D digital twins. The wide adoption of such models has been primarily enabled by robust, model-based, and automatic 3D reconstruction methods. However, these methods impose requirements on the reconstruction, mainly restricting the solution space to several model types and relying on accurate 2D footprints. Recent research shows that deep-learning-based methods promise highly generic solution space and are footprint-free. Yet, the current training and test datasets are limited, hindering the methods’ development. In this work, we analyze the ubiquity of already existing, open 3D city model datasets and their potential to serve as a large-scale training and test set for 3D reconstruction, where 27 potential dataset collections have been identified. Our review shows that more than 215 million building models are readily available. We firmly believe that this review will facilitate further research on robust automatic 3D city model reconstruction and serve as a reference for benchmarking 3D city models.

Cold Start Latency in Serverless Computing: A Systematic Review, Taxonomy, and Future Directions

Recently, academics and the corporate sector have paid attention to serverless computing, which enables dynamic scalability and an economic model. In serverless computing, users only pay for the time they actually use resources, enabling zero scaling to optimise cost and resource utilisation. However, this approach also introduces the serverless cold start problem. Researchers have developed various solutions to address the cold start problem, yet it remains an unresolved research area. In this article, we propose a systematic literature review on clod start latency in serverless computing. Furthermore, we create a detailed taxonomy of approaches to cold start latency, which we use to investigate existing techniques for reducing the cold start time and frequency. We have classified the current studies on cold start latency into several categories such as caching and application-level optimisation-based solutions, as well as Artificial Intelligence (AI)/Machine Learning (ML)-based solutions. Moreover, we have analyzed the impact of cold start latency on quality of service, explored current cold start latency mitigation methods, datasets, and implementation platforms, and classified them into categories based on their common characteristics and features. Finally, we outline the open challenges and highlight the possible future directions.

Study of the patterns of ice lake variation and the factors influencing these changes in the western Nyingchi area

Abstract The current ice lake dataset in the western region of Nyingchi requires further improvement. Due to the intricate distribution of ice lakes and imprecise boundary delineation, research tends to overlook small-scale ice lakes in this area. Moreover, most related studies have focused solely on variations in ice lake areas within key regions, such as the Himalayas, with little attention given to changes occurring in southeastern Tibet. The frequency of ice and snow disasters in the study area has been steadily increasing over the years. Therefore, this study utilizes Landsat satellite images and employs visual interpretation methods to generate more precise and comprehensive maps depicting the distribution of ice lakes in the western region of Nyingchi Province for the years 1994, 2010, 2018, and 2022. Additionally, changes in scale and spatial patterns of different types of ice lakes were investigated. Between 1994 and 2022, the ice lake area in the study area significantly increased by 22.5%, reaching a total of 35.8 ± 3.0 km2. This expansion was primarily driven by glacier-fed lakes, which experienced a remarkable growth rate of 30.8%. In contrast, the non-glacier-fed lakes experienced an increase by only 15.6%. Notably, ice lakes at higher elevations exhibited a peak in expansion, with those above 5143.0 m experiencing the most substantial growth rate of 44.8%. The long-term expansion rate of ice lakes is investigated through the measurement of changes in their boundaries, with the aim to understand the factors contributing to their growth. These findings indicate the rapid expansion of the ice lake near the glacier, with an annual growth rate of 1.3% per annum. Specifically, the glacial-fed section exhibited an expansion rate of 1.1% per annum, while the nonglacial-fed section experienced a growth rate of 0.6% per annum. The seasonal variability in marine glaciers is the primary factor influencing the expansion of ice lakes in this region, with temperature and precipitation serving as the principal driving forces impacting the transformation of these lakes. The data provided by the research results will facilitate a comprehensive understanding of the dynamics and mechanisms governing the ice lake in western Nyingchi, thereby contributing to an enhanced scientific comprehension of potential disaster risks associated with this ice lake.

Developing of a multi-angle and varied walking conditions dataset for human gait recognition

The field of Human Gait Recognition (HGR) leverages unique walking patterns for non-invasive, discreet biometric identification. This study highlights the importance of comprehensive datasets in the development, testing, and validation of HGR algorithms. While current datasets, such as those from Carnegie Mellon University Motion of Body (CMU MoBo), Southampton (SOTON), Chinese Academic of Science Institute of Automation (CASIA B), and Osaka University Institute of Scientific Industrial Research (OU-ISIR), have advanced in scale and complexity, they often lack diversity and comprehensive sample representation. To address this gap, we introduce the TecNM Gait-DS dataset (Tecnologico Nacional de Mexico), specifically designed for Latin American populations, featuring 13 viewing angles and five walking variations. Utilizing a Self-Supervised Vision Transformer DINO (Deeper Into Neural Networks) model for view angle classification, our evaluation demonstrates significant improvements in classification accuracy. This dataset not only enhances sample diversity but also supports the development of more robust HGR systems. Our results underscore the potential for improved accuracy and ethical considerations in HGR, advocating for ongoing refinement of datasets to achieve optimal performance and societal acceptance.

Assessing human-source microbial contamination of stormwater in the U.S

Stormwater capture and use (SCU) projects have the potential to provide a significant portion of municipal water demand. However, uncertainty about the degree of microbial contamination in stormwater and the required treatment is a barrier for implementation of SCU projects. Stormwater runoff could become contaminated with human fecal matter in areas with deteriorating infrastructure where raw wastewater exfiltrates from sewer networks to stormwater collection networks, homeless encampments exist, or sanitary sewer overflows (SSOs) occur. Estimation of human fecal contamination can inform selection of stormwater treatment targets. This study investigates stormwater microbial contamination originating from human fecal matter using observed detections and concentrations of human microbial source tracking (MST) markers and potentially human-infectious pathogens (PHIPs). First, a systematic review complied measurements of human MST markers in wet and dry weather stormwater flows and influent wastewater. In addition, measurements of viral pathogens (e.g., adenoviruses, norovirus GI+GII, and enteroviruses) and protozoan pathogens (e.g., Giardia lamblia and Cryptosporidium parvum) in wet weather flows and influent wastewater were assessed. Human MST marker and PHIP data were statistically analyzed and applied to estimate a human fecal contamination analog (HFCA) which is an estimate of the amount of human fecal matter based on relative concentrations of microbial contaminants in stormwater compared to municipal wastewater. Human MST-based HFCAs in wet and dry weather flows ranged from <10−7.0 to 10−1.5 (median = 10−4.5) and 10−12 to 10−2.6 (median = 10−7.0), respectively. PHIP-based HFCAs in wet weather flows ranged from ∼10−8 to 10−0.14. Estimates of human MST-based HFCAs are more reliable than PHIP-based HFCAs because the current PHIP datasets are generally limited by the number of data points, percent detection, variability observed within the statistical distributions, and the geographical span of sampling locations. The use of human MST-based HFCAs is recommended to guide the selection of stormwater treatment process trains that are protective of public health based on the intended end use. Application of HFCA 10−1 (i.e., sewage dilution 10−1) remains a reasonable conservative estimate of human fecal contamination in stormwater to inform selection of pathogen log reduction targets based on the data presently available.

Characteristics of trauma patients treated by Helicopter Emergency Medical Service and transported to the hospital by helicopter or ambulance

IntroductionTrauma patients treated by the Helicopter Emergency Medical Services (HEMS) can be transported to the hospital either by helicopter or by ambulance, in both cases accompanied by the HEMS physician. The objectives of this study are first to compile an overview of patients treated and transported by the HEMS team with either the helicopter (patients transported by helicopter, PTH) or with the ambulance (patients transported by ambulance, PTA). In addition, to evaluate whether the existing information systems obtain relevant data for researching the decision-making process. The second objective is to identify potentially influencing factors that could be significant for further research.MethodsAll patients in the period from 1 January 2011 until 31 December 2020, treated by HEMS and subsequently transported to hospitals were included in the study. To avoid overrepresentation of the PTA group, a random sample was taken, creating two groups in a 1:2 ratio (PTH n = 724, PTA n = 1448). Differences in patient and treatment characteristics between PTH and PTA were compared using t-tests, Mann-Whitney U tests, and chi-square tests.ResultsPTH accounted for 12.2% of all transports. Approximately two-third of the patients were male and the mean age was around 40 years. PTH had lower iEMV (initial Eye opening, best Motor response, best Verbal response) and iRTS (initial Revised Trauma Score) scores, were more frequently transported to a level 1 trauma centre, underwent more prehospital treatments and were roughly twice as far from their hospital of arrival compared to PTA.ConclusionThe current dataset is, after some modifications, suitable to provide a comprehensive overview of patients treated by HEMS in the Netherlands. A predictive model could be developed using this dataset, which should include factors such as the patient’s location, age, distance to the hospital, physician on duty, mechanism of injury and overall injury severity.

Wave overtopping discharges at rubble mound structures in shallow water

Wave overtopping of coastal structures has been studied using physical model experiments with rubble mound breakwaters in shallow water. The mean overtopping discharge is determined for three different foreshore slopes and various hydrodynamic conditions. The hydrodynamic results confirm that energy is transferred to low-frequency waves in very shallow water and that the short waves are in phase with the lower-frequency waves in very shallow water. As a result, the extreme waves (e.g. 2% exceedance wave height) become relatively large in very shallow water due to the energy of the low-frequency waves affecting thereby the wave overtopping. To estimate the amount of energy at the low-frequency waves, an expression is derived which reasonably accurately predicts the low-frequency wave energy (RMSE of 0.06). Considering the non-dimensional overtopping discharge, the existing formulations for the non-dimensional mean wave overtopping discharge perform poorly to reasonably in shallow water with RMSLE ranging from 1.04 to 2.92. A parameter sensitivity study shows that the short-wave steepness, relative crest height and the low-frequency wave height are the most important parameters when predicting the mean overtopping discharge in shallow water. When including the short-wave steepness and relative crest height in an empirical formulation the RMSLE for the current dataset reduces to 0.69. A further increase in accuracy is found when the low-frequency wave height and 2% exceedance wave height are included (RMSLE 0.64).

Toward Attribute-Controlled Fashion Image Captioning

Fashion image captioning is a critical task in the fashion industry that aims to automatically generate product descriptions for fashion items. However, existing fashion image captioning models predict a fixed caption for a particular fashion item once deployed, which does not cater to unique preferences. We explore a controllable way of fashion image captioning that allows the users to specify a few semantic attributes to guide the caption generation. Our approach utilizes semantic attributes as a control signal, giving users the ability to specify particular fashion attributes (e.g., stitch, knit, sleeve) and styles (e.g., cool, classic, fresh) that they want the model to incorporate when generating captions. By providing this level of customization, our approach creates more personalized and targeted captions that suit individual preferences. To evaluate the effectiveness of our proposed approach, we clean, filter, and assemble a new fashion image caption dataset called FACAD170K from the current FACAD dataset. This dataset facilitates learning and enables us to investigate the effectiveness of our approach. Our results demonstrate that our proposed approach outperforms existing fashion image captioning models as well as conventional captioning methods. Besides, we further validate the effectiveness of the proposed method on the MSCOCO and Flickr30K captioning datasets and achieve competitive performance.

The Effect of Probiotics and Microbiota on Nervous System and Mental Illnesses.

The microbiota that inhabits the gastrointestinal tract has been linked to various gastrointestinal and non-gastrointestinal disorders. Scientists have been studying how the bacteria in our intestines have an effect on our brain and nervous system. This connection is called the "microbiota- gut-brain axis". Given the capacity of probiotics, which are live non-pathogenic microorganisms, to reinstate the normal microbial population within the host and confer advantages, their potential impact has been subjected to scrutiny with regard to neurological and mental conditions. Material sourced for this review included peer-reviewed literature annotated in the PubMed, Web of Science, Scopus, and Google Scholar databases. The result has indicated the integration of probiotics into a child's diet to enhance the neuro-behavioral symptoms. Notwithstanding this, the current data set has been found to be insufficient and inconclusive. The potential utility of probiotics for the prevention or treatment of neurologic and mental disorders has become a subject of substantial interest.

Modeling and Optimization of Interaction Behavior between Right Turning Motor Vehicles and Pedestrians at Signalized Intersections based on Machine Learning

This project first conducted a comprehensive analysis of the interaction process between pedestrians and right turning motor vehicles and its influencing factors through a summary of existing literature and observation of actual situations. On the basis of comprehensive and accurate data, machine learning algorithms were used to model and predict the interaction behavior between pedestrians and right turning motor vehicles. By comparing different algorithms, the optimal models for predicting the yielding behavior of right turning motor vehicles and pedestrian crossing behavior in the current dataset were obtained. According to the model evaluation report, the obtained models have better predictive performance on the test set. And based on the obtained optimal model, the feature importance ranking was carried out for the pedestrian crossing model and the right turn motor vehicle yielding model, and the importance scores of the influencing factors of pedestrian crossing behavior and right turn motor vehicle yielding behavior were obtained respectively. Finally, a comparison was made between the modeling effects of the crossing gap feature group and the speed distance feature group, further clarifying the importance of the speed distance index for right turning motor vehicles in predicting interactive behavior.

Linking transcriptome and morphology in bone cells at cellular resolution with generative AI

Abstract Recent advancements in deep learning (DL) have revolutionized the capability of artificial intelligence (AI) by enabling the analysis of large-scale, complex datasets that are difficult for humans to interpret. However, large amounts of high-quality data are required to train such generative AI models successfully. With the rapid commercialization of single-cell sequencing and spatial transcriptomics platforms, the field is increasingly producing large-scale datasets such as histological images, single-cell molecular data, and spatial transcriptomic data. These molecular and morphological datasets parallel the multimodal text and image data used to train highly successful generative AI models for natural language processing and computer vision. Thus, these emerging data types offer great potential to train generative AI models that uncover intricate biological processes of bone cells at a cellular level. In this Perspective, we summarize the progress and prospects of generative AI applied to these datasets and their potential applications to bone research. In particular, we highlight three AI applications: predicting cell differentiation dynamics, linking molecular and morphological features, and predicting cellular responses to perturbations. To make generative AI models beneficial for bone research, important issues, such as technical biases in bone single-cell datasets, lack of profiling of important bone cell types, and lack of spatial information, need to be addressed. Realizing the potential of generative AI for bone biology will also likely require generating large-scale, high-quality cellular-resolution spatial transcriptomics datasets, improving the sensitivity of current spatial transcriptomics datasets, and thorough experimental validation of model predictions.

Outcomes of Medicaid Rebalancing May Differ Across Enrollee Populations.

Access to home- and community-based services (HCBS) may prevent or delay nursing home transitions among older adults. Medicaid's Balancing Incentive Program (BIP) (2011-2015) provided assistance for U.S. states to increase access to HCBS through infrastructure changes and spending benchmarks. We combined longitudinal data from the 2008-2019 Medicare Current Beneficiary Survey and Minimum Data Set and used survival modeling to examine the association between BIP exposure (living in a BIP-participant state vs. not) and time to long-term institutionalization (LTI, defined as a nursing home episode of 90+ days) among dual enrollees ages 65 and older. In the main effects model, BIP exposure was not associated with hazard of LTI. Interaction models showed that BIP exposure was associated with a lower hazard of LTI among Hispanic/Latinx enrollees, while the opposite was true among non-Hispanic White enrollees. Our findings suggest the outcomes of Medicaid rebalancing efforts may differ across enrollee subgroups.

Gas Detection and Classification Using Multimodal Data Based on Federated Learning.

The identification of gas leakages is a significant factor to be taken into consideration in various industries such as coal mines, chemical industries, etc., as well as in residential applications. In order to reduce damage to the environment as well as human lives, early detection and gas type identification are necessary. The main focus of this paper is multimodal gas data that were obtained simultaneously by using multiple sensors for gas detection and a thermal imaging camera. As the reliability and sensitivity of low-cost sensors are less, they are not suitable for gas detection over long distances. In order to overcome the drawbacks of relying just on sensors to identify gases, a thermal camera capable of detecting temperature changes is also used in the collection of the current multimodal dataset The multimodal dataset comprises 6400 samples, including smoke, perfume, a combination of both, and neutral environments. In this paper, convolutional neural networks (CNNs) are trained on thermal image data, utilizing variants such as bidirectional long-short-term memory (Bi-LSTM), dense LSTM, and a fusion of both datasets to effectively classify comma separated value (CSV) data from gas sensors. The dataset can be used as a valuable source for research scholars and system developers to improvise their artificial intelligence (AI) models used for gas leakage detection. Furthermore, in order to ensure the privacy of the client's data, this paper explores the implementation of federated learning for privacy-protected gas leakage classification, demonstrating comparable accuracy to traditional deep learning approaches.

Internal validation of modified Mirels’ scoring system for pathologic femur fractures

BackgroundThe proximal femur is a common site of bone metastasis. The Mirels’ score is a frequently utilized system to identify patients at risk for pathologic fracture and while it has consistently demonstrated strong sensitivity, specificity has been relatively poor. Our group previously developed a Modified Mirels’ scoring system which demonstrated improved ability to predict cases at risk of fracture in this patient population through modification of the Mirels’ location score. The purpose of the present study is to internally validate this newly developed scoring system on an independent patient series.MethodsRetrospective review was performed to identify patients who were evaluated for proximal femoral bone lesions. Patients were stratified into one of two groups: 1) those who went on to fracture within 4 months after initial evaluation (Fracture Group) and 2) those who did not fracture within 4 months of initial evaluation (No Fracture Group). Retrospective chart review was performed to assign an Original Mirels’ (OM) Score and Modified Mirels’ (MM) score to each patient at the time of initial evaluation. Descriptive statistics, logistic regression, receiver operating curve, and net benefit analyses were performed to determine the predictability of fractures when utilizing both scoring systems.ResultsThe use of the MM scoring improved fracture prediction over OM scoring for patients observed over a 4 month follow up based on logistic regression. Decision curve analysis showed that there was a net benefit using the MM score over the OM scoring for a full range of fracture threshold probabilities. Fracture prevalence was similar for current internal validation dataset when compared to the dataset of our index study with a comparable reduction in misclassification of fracture prediction when utilizing the modified scoring system versus the original.ConclusionsUse of MM scoring was found to improve fracture prediction over OM scoring when tested on an internal validation set of patients with disseminated metastatic lesions to the proximal femur. The improvement in fracture prediction demonstrated in the present study mirrored the results of our index study during which the MM system was developed.

Utilizing a Dashboard for Efficient Biodiversity Data Overview and Gap-Filling Effort

Taiwan Biodiversity Information Facility has been leading the effort to integrate and mobilize biodiversity data nationally. We recently accomplished the integration of local databases and successfully developed a data portal that efficiently links to the most comprehensive biodiversity database in Taiwan: the Taiwan Biodiversity Information Alliance Data Portal. However, we identified an emerging challenge following this development. Preliminary analyses have revealed several gaps in the current dataset, impeding robust data use. For instance, land data dominates over sea data, and bird records are more abundant than those of other taxonomic groups. In our endeavour to enhance national biodiversity data coverage, we are actively exploring methodologies to address these existing data gaps. Here, we broadly define data gap as the absence of required data or information for use. Our primary objective is to empower users to identify these gaps, thereby facilitating efficient, informed survey planning and eventually filling these gaps. For this purpose, we are evaluating the effectiveness of a user-friendly dashboard, tailored for biodiversity survey planning in Taiwan. We developed and deployed a R Shiny dashboard*1 that provides users with insightful data visualizations that highlight spatial, temporal, and taxonomical data overviews. It also offers customizable filtering options, such as by major taxon groups and habitat types, for easy viewing. Additionally, the dashboard includes a download function, so users can access spatial and taxon gap information in their survey planning efforts. After deploying the dashboard, we conducted user interviews to collect feedback aimed at refining its usability and effectiveness. In addition to helping survey planning, we anticipate that this tool will also aid data users in promptly assessing data integrity within the national biodiversity data portal, thereby enabling them to identify potential data gaps and biases that may affect subsequent use. Through this presentation, we aim to share our experience in developing an interactive dashboard tailored for public use. Preliminary results had indicated that the dashboard highlighted some visible data gaps and biases, particularly in underrepresented areas such as marine habitats and less-studied taxonomic groups. Additionally, user interviews revealed a demand for a feature that provides single species data overviews, suggesting a potential area for further development to enhance the tool's utility. Some challenges remain, such as the need to enhance data granularity and user interface usability. Future work will focus on incorporating more real-time data integration and expanding the range of customizable filters. We believe these efforts will further empower users in Taiwan to conduct more informed and efficient biodiversity surveys, ultimately supporting research and policy-making to address critical biodiversity data gaps.

A national Australian Congenital Heart Disease registry; methods and initial results

BackgroundAlthough several National Data Registries for Congenital Heart Disease (CHD) exist, few are comprehensive and contemporary. A National Australian CHD Registry has been developed that aims to redress this by creating the first comprehensive data collection for CHD children and adults, initially across Australia. MethodsWe defined and collected a minimum dataset of demographics, diagnoses, and procedures from people with CHD presenting at participating quaternary CHD services Australia-wide. Data were collected from a range of clinical data sources. Diagnoses and procedures were standardised to the European Paediatric Congenital Code – Short List. Methodological limitations were carefully documented. ResultsFrom 8 participating institutions, an initial 359,084 patient records were assessed for eligibility and 68,234 unique individuals with structural CHD have been included in the current dataset. There were 20,395 (30 %) people with mild CHD, 25,157 (37 %) with moderate CHD, and 13,530 (20 %) with severe CHD (6 % unknown complexity). The most common diagnoses were Ventricular Septal Defect (16,781, 25 %), Atrial Septal Defect (6,607, 10 %), Aortic Valve Disorders (5516 8 %), Coarctation of the Aorta (5,321, 8 %), Tetralogy of Fallot (4,489, 7 %), Transposition of the Great Arteries (4,009, 6 %). ConclusionThe data presented here represents the most comprehensive cohort collected for the Australian CHD population thus far and is comparable with the largest contemporary CHD registries around the world. This Registry represents a key resource for improved understanding of the CHD population and will drive better care and outcomes for people living with CHD.

Network Attack Classification with a Shallow Neural Network for Internet and Internet of Things (IoT) Traffic

In recent years, there has been a tremendous increase in the use of connected devices as part of the so-called Internet of Things (IoT), both in private spaces and the industry. Integrated distributed systems have shown many benefits compared to isolated devices. However, exposing industrial infrastructure to the global Internet also generates security challenges that need to be addressed to benefit from tighter systems integration and reduced reaction times. Machine learning algorithms have demonstrated their capacity to detect sophisticated cyber attack patterns. However, they often consume significant amounts of memory, computing resources, and scarce energy. Furthermore, their training relies on the availability of datasets that accurately represent real-world data traffic subject to cyber attacks. Network attacks are relatively rare events, as is reflected in the distribution of typical training datasets. Such imbalanced datasets can bias the training of a neural network and prevent it from successfully detecting underrepresented attack samples, generally known as the problem of imbalanced learning. This paper presents a shallow neural network comprising only 110 ReLU-activated artificial neurons capable of detecting representative attacks observed on a communication network. To enable the training of such small neural networks, we propose an improved attack-sharing loss function to cope with imbalanced learning. We demonstrate that our proposed solution can detect network attacks with an F1 score above 99% for various attacks found in current intrusion detection system datasets, focusing on IoT device communication. We further show that our solution can reduce the false negative detection rate of our proposed shallow network and thus further improve network security while enabling processing at line rate in low-complexity network intrusion systems.

Acquisition of Data on Kinematic Responses to Unpredictable Gait Perturbations: Collection and Quality Assurance of Data for Use in Machine Learning Algorithms for (Near-)Fall Detection.

Slip, trip, and fall (STF) accidents cause high rates of absence from work in many companies. During the 2022 reporting period, the German Social Accident Insurance recorded 165,420 STF accidents, of which 12 were fatal and 2485 led to disability pensions. Particularly in the traffic, transport and logistics sector, STF accidents are the most frequently reported occupational accidents. Therefore, an accurate detection of near-falls is critical to improve worker safety. Efficient detection algorithms are essential for this, but their performance heavily depends on large, well-curated datasets. However, there are drawbacks to current datasets, including small sample sizes, an emphasis on older demographics, and a reliance on simulated rather than real data. In this paper we report the collection of a standardised kinematic STF dataset from real-world STF events affecting parcel delivery workers and steelworkers. We further discuss the use of the data to evaluate dynamic stability control during locomotion for machine learning and build a standardised database. We present the data collection, discuss the classification of the data, present the totality of the data statistically, and compare it with existing databases. A significant research gap is the limited number of participants and focus on older populations in previous studies, as well as the reliance on simulated rather than real-world data. Our study addresses these gaps by providing a larger dataset of real-world STF events from a working population with physically demanding jobs. The population studied included 110 participants, consisting of 55 parcel delivery drivers and 55 steelworkers, both male and female, aged between 19 and 63 years. This diverse participant base allows for a more comprehensive understanding of STF incidents in different working environments.