Abstract

The dataset is a relational dataset of 8,000 households households, representing a sample of the population of an imaginary middle-income country. The dataset contains two data files: one with variables at the household level, the other one with variables at the individual level. It includes variables that are typically collected in population censuses (demography, education, occupation, dwelling characteristics, fertility, mortality, and migration) and in household surveys (household expenditure, anthropometric data for children, assets ownership). The data only includes ordinary households (no community households). The dataset was created using REaLTabFormer, a model that leverages deep learning methods. The dataset was created for the purpose of training and simulation and is not intended to be representative of any specific country.

The full-population dataset (with about 10 million individuals) is also distributed as open data.

Geographic coverage

The dataset is a synthetic dataset for an imaginary country. It was created to represent the population of this country by province (equivalent to admin1) and by urban/rural areas of residence.

Analysis unit

Household, Individual

Universe

The dataset is a fully-synthetic dataset representative of the resident population of ordinary households for an imaginary middle-income country.

Kind of data

ssd

Sampling procedure

The sample size was set to 8,000 households. The fixed number of households to be selected from each enumeration area was set to 25. In a first stage, the number of enumeration areas to be selected in each stratum was calculated, proportional to the size of each stratum (stratification by geo_1 and urban/rural). Then 25 households were randomly selected within each enumeration area. The R script used to draw the sample is provided as an external resource.

Mode of data collection

other

Research instrument

The dataset is a synthetic dataset. Although the variables it contains are variables typically collected from sample surveys or population censuses, no questionnaire is available for this dataset. A "fake" questionnaire was however created for the sample dataset extracted from this dataset, to be used as training material.

Cleaning operations

The synthetic data generation process included a set of "validators" (consistency checks, based on which synthetic observation were assessed and rejected/replaced when needed). Also, some post-processing was applied to the data to result in the distributed data files.

Response rate

This is a synthetic dataset; the "response rate" is 100%.

This repository stores synthetic datasets derived from the database of the UK Biobank (UKB) cohort.

The datasets were generated for illustrative purposes, in particular for reproducing specific analyses on the health risks associated with long-term exposure to air pollution using the UKB cohort. The code used to create the synthetic datasets is available and documented in a related GitHub repo, with details provided in the section below. These datasets can be freely used for code testing and for illustrating other examples of analyses on the UKB cohort.

The synthetic data have been used so far in two analyses described in related peer-reviewed publications, which also provide information about the original data sources:

• Vanoli J, et al. Long-term associations between time-varying exposure to ambient PM2.5 and mortality: an analysis of the UK Biobank. Epidemiology. 2025;36(1):1-10. DOI: 10.1097/EDE.0000000000001796 [freely available here, with code provided in this GitHub repo]
• Vanoli J, et al. Confounding issues in air pollution epidemiology: an empirical assessment with the UK Biobank cohort. International Journal of Epidemiology. 2025;54(5):dyaf163. DOI: 10.1093/ije/dyaf163 [freely available here, with code provided in this GitHub repo]

Note: while the synthetic versions of the datasets resemble the real ones in several aspects, the users should be aware that these data are fake and must not be used for testing and making inferences on specific research hypotheses. Even more importantly, these data cannot be considered a reliable description of the original UKB data, and they must not be presented as such.

The work was supported by the Medical Research Council-UK (Grant ID: MR/Y003330/1).

Content

The series of synthetic datasets (stored in two versions with csv and RDS formats) are the following:

• synthbdcohortinfo: basic cohort information regarding the follow-up period and birth/death dates for 502,360 participants.
• synthbdbasevar: baseline variables, mostly collected at recruitment.
• synthpmdata: annual average exposure to PM_2.5 for each participant reconstructed using their residential history.
• synthoutdeath: death records that occurred during the follow-up with date and ICD-10 code.

In addition, this repository provides these additional files:

• codebook: a pdf file with a codebook for the variables of the various datasets, including references to the fields of the original UKB database.
• asscentre: a csv file with information on the assessment centres used for recruitment of the UKB participants, including code, names, and location (as northing/easting coordinates of the British National Grid).
• Countries_December_2022_GB_BUC: a zip file including the shapefile defining the boundaries of the countries in Great Britain (England, Wales, and Scotland), used for mapping purposes [source].

Generation of the synthetic data

The datasets resemble the real data used in the analysis, and they were generated using the R package synthpop (www.synthpop.org.uk). The generation process involves two steps, namely the synthesis of the main data (cohort info, baseline variables, annual PM_2.5 exposure) and then the sampling of death events. The R scripts for performing the data synthesis are provided in the GitHub repo (subfolder Rcode/synthcode).

The first part merges all the data, including the annual PM_2.5 levels, into a single wide-format dataset (with a row for each subject), generates a synthetic version, adds fake IDs, and then extracts (and reshapes) the single datasets. In the second part, a Cox proportional hazard model is fitted on the original data to estimate risks associated with various predictors (including the main exposure represented by PM_2.5), and then these relationships are used to simulate death events in each year. Details on the modelling aspects are provided in the article.

This process guarantees that the synthetic data do not hold specific information about the original records, thus preserving confidentiality. At the same time, the multivariate distribution and correlation across variables, as well as the mortality risks, resemble those of the original data, so the results of descriptive and inferential analyses are similar to those in the original assessments. However, as noted above, the data are used only for illustrative purposes, and they must not be used to test other research hypotheses.

These are simulated data without any identifying information or informative birth-level covariates. We also standardize the pollution exposures on each week by subtracting off the median exposure amount on a given week and dividing by the interquartile range (IQR) (as in the actual application to the true NC birth records data). The dataset that we provide includes weekly average pregnancy exposures that have already been standardized in this way while the medians and IQRs are not given. This further protects identifiability of the spatial locations used in the analysis. This dataset is not publicly accessible because: EPA cannot release personally identifiable information regarding living individuals, according to the Privacy Act and the Freedom of Information Act (FOIA). This dataset contains information about human research subjects. Because there is potential to identify individual participants and disclose personal information, either alone or in combination with other datasets, individual level data are not appropriate to post for public access. Restricted access may be granted to authorized persons by contacting the party listed. It can be accessed through the following means: File format: R workspace file; “Simulated_Dataset.RData”. Metadata (including data dictionary) • y: Vector of binary responses (1: adverse outcome, 0: control) • x: Matrix of covariates; one row for each simulated individual • z: Matrix of standardized pollution exposures • n: Number of simulated individuals • m: Number of exposure time periods (e.g., weeks of pregnancy) • p: Number of columns in the covariate design matrix • alpha_true: Vector of “true” critical window locations/magnitudes (i.e., the ground truth that we want to estimate) Code Abstract We provide R statistical software code (“CWVS_LMC.txt”) to fit the linear model of coregionalization (LMC) version of the Critical Window Variable Selection (CWVS) method developed in the manuscript. We also provide R code (“Results_Summary.txt”) to summarize/plot the estimated critical windows and posterior marginal inclusion probabilities. Description “CWVS_LMC.txt”: This code is delivered to the user in the form of a .txt file that contains R statistical software code. Once the “Simulated_Dataset.RData” workspace has been loaded into R, the text in the file can be used to identify/estimate critical windows of susceptibility and posterior marginal inclusion probabilities. “Results_Summary.txt”: This code is also delivered to the user in the form of a .txt file that contains R statistical software code. Once the “CWVS_LMC.txt” code is applied to the simulated dataset and the program has completed, this code can be used to summarize and plot the identified/estimated critical windows and posterior marginal inclusion probabilities (similar to the plots shown in the manuscript). Optional Information (complete as necessary) Required R packages: • For running “CWVS_LMC.txt”: • msm: Sampling from the truncated normal distribution • mnormt: Sampling from the multivariate normal distribution • BayesLogit: Sampling from the Polya-Gamma distribution • For running “Results_Summary.txt”: • plotrix: Plotting the posterior means and credible intervals Instructions for Use Reproducibility (Mandatory) What can be reproduced: The data and code can be used to identify/estimate critical windows from one of the actual simulated datasets generated under setting E4 from the presented simulation study. How to use the information: • Load the “Simulated_Dataset.RData” workspace • Run the code contained in “CWVS_LMC.txt” • Once the “CWVS_LMC.txt” code is complete, run “Results_Summary.txt”. Format: Below is the replication procedure for the attached data set for the portion of the analyses using a simulated data set: Data The data used in the application section of the manuscript consist of geocoded birth records from the North Carolina State Center for Health Statistics, 2005-2008. In the simulation study section of the manuscript, we simulate synthetic data that closely match some of the key features of the birth certificate data while maintaining confidentiality of any actual pregnant women. Availability Due to the highly sensitive and identifying information contained in the birth certificate data (including latitude/longitude and address of residence at delivery), we are unable to make the data from the application section publically available. However, we will make one of the simulated datasets available for any reader interested in applying the method to realistic simulated birth records data. This will also allow the user to become familiar with the required inputs of the model, how the data should be structured, and what type of output is obtained. While we cannot provide the application data here, access to the North Carolina birth records can be requested through the North Carolina State Center for Health Statistics, and requires an appropriate data use agreement. Description Permissions: These are simulated data without any identifying information or informative birth-level covariates. We also standardize the pollution exposures on each week by subtracting off the median exposure amount on a given week and dividing by the interquartile range (IQR) (as in the actual application to the true NC birth records data). The dataset that we provide includes weekly average pregnancy exposures that have already been standardized in this way while the medians and IQRs are not given. This further protects identifiability of the spatial locations used in the analysis. This dataset is associated with the following publication: Warren, J., W. Kong, T. Luben, and H. Chang. Critical Window Variable Selection: Estimating the Impact of Air Pollution on Very Preterm Birth. Biostatistics. Oxford University Press, OXFORD, UK, 1-30, (2019).

Abstract

Developing robust deep learning models for fetal ultrasound image analysis requires comprehensive, high-quality datasets to effectively learn informative data representations within the domain. However, the scarcity of labelled ultrasound images poses substantial challenges, especially in low-resource settings. To tackle this challenge, we leverage synthetic data to enhance the generalizability of deep learning models. This study proposes a diffusion-based method, Fetal Ultrasound LoRA (FU-LoRA), which involves fine-tuning latent diffusion models using the LoRA technique to generate synthetic fetal ultrasound images. These synthetic images are integrated into a hybrid dataset that combines real-world and synthetic images to improve the performance of zero-shot classifiers in low-resource settings. Our experimental results on fetal ultrasound images from African cohorts demonstrate that FU-LoRA outperforms the baseline method by a 13.73% increase in zero-shot classification accuracy. Furthermore, FU-LoRA achieves the highest accuracy of 82.40%, the highest F-score of 86.54%, and the highest AUC of 89.78%. It demonstrates that the FU-LoRA method is effective in the zero-shot classification of fetal ultrasound images in low-resource settings. Our code and data are publicly accessible on GitHub.

Method

Our FU-LoRA method: Fine-tuning the pre-trained latent diffusion model (LDM) [2] using the LoRA method on a small fetal ultrasound dataset from high-resource settings (HRS). This approach integrates synthetic images to enhance generalization and performance of deep learning models. We conduct three fine-tuning sessions for the diffusion model to generate three LoRA models with different hyper-parameters: alpha in [8, 32, 128], and r in [8, 32, 128]. The merging rate alpha/r is fixed to 1. The purpose of this operation is to delve deeper into LoRA to uncover optimizations that can improve the model's performance and evaluate the effectiveness of parameter r in generating synthetic images.

Datasets

The Spanish dataset (URL) in HRS includes 1,792 patient records in Spain [1]. All images are acquired during screening in the second and third trimesters of pregnancy using six different machines operated by operators with similar expertise. We randomly selected 20 Spanish ultrasound images from each of the five maternal–fetal planes (Abdomen, Brain, Femur, Thorax, and Other) to fine-tune the LDM using LoRA technique, and 1150 Spanish images (230 x 5 planes) to create the hybrid dataset. In summary, fine-tuning the LDM utilizes 100 images including 85 patients. Training downstream classifiers uses 6148 images from 612 patients. Within the 6148 images used for training, a subset of 200 images is randomly selected for validation purposes. The hybrid dataset employed in this study has a total of 1150 Spanish images, representing 486 patients.

We create the synthetic dataset comprising 5000 fetal ultrasound images (500 x 2 samplers x 5 planes) accessible to the open-source community. The generation process utilizes our LoRA model Rank r = 128 with Euler and UniPC samplers known for their efficiency. Subsequently, we integrate this synthetic dataset with a small amount of Spanish data to create a hybrid dataset.

Implementation Details

The hyper-parameters of LoRA models are defined as follows: batch size to 2; LoRA learning rate to 1e-4; total training steps to 10000; LoRA dimension to 128; mixed precision selection to fp16; learning scheduler to constant; and input size (resolution) to 512. The model is trained on a single NVIDIA RTX A5000, 24 GB with 8-bit Adam optimizer on PyTorch.

The dataset used in this study is publicly available for research purposes. If you are using this dataset, please cite the following paper, which outlines the complete details of the dataset and the methodology used for its generation:

Amit Karamchandani, Javier Núñez, Luis de-la-Cal, Yenny Moreno, Alberto Mozo, Antonio Pastor, "On the Applicability of Network Digital Twins in Generating Synthetic Data for Heavy Hitter Discrimination," under submission.

This dataset contains a synthetic dataset generated to differentiate between benign and malicious heavy hitter flows within complex network environments. Heavy Hitter flows, which include high-volume data transfers, can significantly impact network performance, leading to congestion and degraded quality of service. Distinguishing legitimate heavy hitter activity from malicious Distributed Denial-of-Service traffic is critical for network management and security, yet existing datasets lack the granularity needed for training machine learning models to effectively make this distinction.

To address this, a Network Digital Twin (NDT) approach was utilized to emulate realistic network conditions and traffic patterns, enabling automated generation of labeled data for both benign and malicious HH flows alongside regular traffic.

The feature set includes flow statistics commonly used in network analysis, such as:

• Traffic protocol type,
• Flow duration (the time between the initial and final packet in both directions),
• Total count of payload packets transmitted in both directions,
• Cumulative bytes transmitted in both directions,
• Time discrepancy between the first packet observations at the source and destination,
• Packet and byte transmission rates per second within each interval, and
• Total packet and byte counts within each interval in both directions.

This synthetic dataset contains 5,000 student records exploring the relationship between study hours and academic performance.

Dataset Features

• student_id: Unique identifier for each student (1-5000)
• study_hours: Hours spent studying (0-12 hours, continuous)
• grade: Final exam score (0-100 points, continuous)

Potential Use Cases

• Linear regression modeling and practice
• Data visualization exercises
• Statistical analysis tutorials
• Machine learning for beginners
• Educational research simulations

Data Quality

• No missing values
• Normally distributed residuals
• Realistic educational scenario
• Ready for immediate analysis

Data Generation Code

This dataset was generated using R.

R Code

# Set seed for reproducibility
set.seed(42)

# Define number of observations (students)
n <- 5000

# Generate study hours (independent variable)
# Uniform distribution between 0 and 12 hours
study_hours <- runif(n, min = 0, max = 12)

# Create relationship between study hours and grade
# Base grade: 40 points
# Each study hour adds an average of 5 points
# Add normal noise (standard deviation = 10)
theoretical_grade <- 40 + 5 * study_hours

# Add normal noise to make it realistic
noise <- rnorm(n, mean = 0, sd = 10)

# Calculate final grade
grade <- theoretical_grade + noise

# Limit grades between 0 and 100
grade <- pmin(pmax(grade, 0), 100)

# Create the dataframe
dataset <- data.frame(
 student_id = 1:n,
 study_hours = round(study_hours, 2),
 grade = round(grade, 2)
)

We include a description of the data sets in the meta-data as well as sample code and results from a simulated data set. This dataset is not publicly accessible because: EPA cannot release personally identifiable information regarding living individuals, according to the Privacy Act and the Freedom of Information Act (FOIA). This dataset contains information about human research subjects. Because there is potential to identify individual participants and disclose personal information, either alone or in combination with other datasets, individual level data are not appropriate to post for public access. Restricted access may be granted to authorized persons by contacting the party listed. It can be accessed through the following means: The R code is available on line here: https://github.com/warrenjl/SpGPCW. Format: Abstract The data used in the application section of the manuscript consist of geocoded birth records from the North Carolina State Center for Health Statistics, 2005-2008. In the simulation study section of the manuscript, we simulate synthetic data that closely match some of the key features of the birth certificate data while maintaining confidentiality of any actual pregnant women. Availability Due to the highly sensitive and identifying information contained in the birth certificate data (including latitude/longitude and address of residence at delivery), we are unable to make the data from the application section publicly available. However, we will make one of the simulated datasets available for any reader interested in applying the method to realistic simulated birth records data. This will also allow the user to become familiar with the required inputs of the model, how the data should be structured, and what type of output is obtained. While we cannot provide the application data here, access to the North Carolina birth records can be requested through the North Carolina State Center for Health Statistics and requires an appropriate data use agreement. Description Permissions: These are simulated data without any identifying information or informative birth-level covariates. We also standardize the pollution exposures on each week by subtracting off the median exposure amount on a given week and dividing by the interquartile range (IQR) (as in the actual application to the true NC birth records data). The dataset that we provide includes weekly average pregnancy exposures that have already been standardized in this way while the medians and IQRs are not given. This further protects identifiability of the spatial locations used in the analysis. File format: R workspace file. Metadata (including data dictionary) • y: Vector of binary responses (1: preterm birth, 0: control) • x: Matrix of covariates; one row for each simulated individual • z: Matrix of standardized pollution exposures • n: Number of simulated individuals • m: Number of exposure time periods (e.g., weeks of pregnancy) • p: Number of columns in the covariate design matrix • alpha_true: Vector of “true” critical window locations/magnitudes (i.e., the ground truth that we want to estimate). This dataset is associated with the following publication: Warren, J., W. Kong, T. Luben, and H. Chang. Critical Window Variable Selection: Estimating the Impact of Air Pollution on Very Preterm Birth. Biostatistics. Oxford University Press, OXFORD, UK, 1-30, (2019).

The aim of this project was to create a synthetic dataset without using the original (secure, controlled) dataset to do so, and instead using only publicly available analytical output (i.e. output that was cleared for publication) to create the synthetic data. Such synthetic data may allow users to gain familiarity with and practise on data that is like the original before they gain access to the original data (where time in a secure setting may be limited).

The Annual Survey of Hours and Earnings 2011 and Census 2011: Synthetic Data was created without access to the original ASHE-2011 Census dataset (which is only available in a secure setting via the ONS Secure Research Service: "Annual Survey of Hours and Earnings linked to 2011 Census - England and Wales"). It was created as a teaching aid to support a training course "An Introduction to the linked ASHE-2011 Census dataset" organised by Administrative Data Research UK and the National Centre for Research Methods. The synthetic dataset contains a subset of the variables in the original dataset and was designed to reproduce the analytical output contained in the ASHE-Census 2011 Data Linkage User Guide.

This is a synthetic dataset that can be used by users that are interested in benchmarking methods of explainable artificial intelligence (XAI) for geoscientific applications. The dataset is specifically inspired from a climate forecasting setting (seasonal timescales) where the task is to predict regional climate variability given global climate information lagged in time. The dataset consists of a synthetic input X (series of 2D arrays of random fields drawn from a multivariate normal distribution) and a synthetic output Y (scalar series) generated by using a nonlinear function F: R^d -> R.

The synthetic input aims to represent temporally independent realizations of anomalous global fields of sea surface temperature, the synthetic output series represents some type of regional climate variability that is of interest (temperature, precipitation totals, etc.) and the function F is a simplification of the climate system.

Since the nonlinear function F that is used to generate the output given the input is known, we also derive and provide the attribution of each output value to the corresponding input features. Using this synthetic dataset users can train any AI model to predict Y given X and then implement XAI methods to interpret it. Based on the “ground truth” of attribution of F the user can assess the faithfulness of any XAI method.

NOTE: the spatial configuration of the observations in the NetCDF database file conform to the planetocentric coordinate system (89.5N - 89.5S, 0.5E - 359.5E), where longitude is measured in the positive heading east from the prime meridian.

This repository contains scripts, input files, and some example output files for the Residential Population Generator, an R-based tool to generate synthetic human residental populations to use in making estimates of near-field chemical exposures. This tool is most readily adapted for using in the workflow for CHEM, the Combined Human Exposure Model, avaialable in two other GitHub repositories in the HumanExposure project, including ProductUseScheduler and source2dose. CHEM is currently best suited to estimating exposure to product use. Outputs from RPGen are translated into ProductUseScheduler, which with subsequent outputs used in source2dose.

Subtitle:

Simulated Dataset of Customer Purchase Behavior

Description:

This dataset contains simulated data representing customer purchase behavior. It includes various features such as age, gender, income, education, region, loyalty status, purchase frequency, purchase amount, product category, promotion usage, and satisfaction score.

File Information:

• File Format: CSV
• Number of Rows: 100000
• Number of Columns: 12

Column Descriptors:

• age: Age of the customer.
• gender: Gender of the customer (0 for Male, 1 for Female).
• income: Annual income of the customer.
• education: Education level of the customer.
• region: Region where the customer resides.
• loyalty_status: Loyalty status of the customer.
• purchase_frequency: Frequency of purchases made by the customer.
• purchase_amount: Amount spent by the customer in each purchase.
• product_category: Category of the purchased product.
• promotion_usage: Indicates whether the customer used promotional offers (0 for No, 1 for Yes).
• satisfaction_score: Satisfaction score of the customer.

Provenance:

The dataset was simulated using the simstudy package in R. Various distributions and formulas were used to generate synthetic data representing customer purchase behavior. The data is organized to mimic real-world scenarios, but it does not represent actual customer data.

The fully synthetic neuroimaging dataset generated for analysis of fully synthetic data in the current study are available as Research Data from Mendeley Data. Ten fully synthetic datasets are included, with synthetic gray matter images (nifti files) that were generated based on simulated participant data (text files). The file Synthetic_predictors.tar.gz contains ten fully synthetic predictor tables with information for 264 simulated subjects. Due to large file sizes, a separate archive was created for each set of synthetic gray matter image data: RBS001.tar.gz, …, RBS010.tar.gz. Regression analyses were performed for each synthetic dataset, then average statistic maps were made for each contrast, which were then smoothed (see accompanying paper for additional information).

The supplementary materials also include commented MATLAB and R code to implement the current neuroimaging data synthesis methods (SKexample.zip). The example data were selected from an earlier fMRI study (Kuchinsky et al., 2012) to demonstrate that the current approach can be used with other types of neuroimaging data. The example code can also be adapted to produce fully synthetic group-level datasets based on observed neuroimaging data from other sources. The zip archive includes a document with important information for performing the example analyses, and details that should be communicated with recipients of a synthetic neuroimaging dataset.

Kuchinsky, S.E., Vaden, K.I., Keren, N.I., Harris, K.C., Ahlstrom, J.B., Dubno, J.R., Eckert, M.A., 2012. Word intelligibility and age predict visual cortex activity during word listening. Cerebral Cortex 22, 1360–71. https://doi.org/10.1093/cercor/bhr211

R code to generate synthetic data. (R 1 kb)

Introduction

Vessel segmentation in fundus images is essential in the diagnosis and prognosis of retinal diseases and the identification of image-based biomarkers. However, creating a vessel segmentation map can be a tedious and time consuming process, requiring careful delineation of the vasculature, which is especially hard for microcapillary plexi in fundus images. Optical coherence tomography angiography (OCT-A) is a relatively novel modality visualizing blood flow and microcapillary plexi not clearly observed in fundus photography. Unfortunately, current commercial OCT-A cameras have various limitations due to their complex optics making them more expensive, less portable, and with a reduced field of view (FOV) compared to fundus cameras. Moreover, the vast majority of population health data collection efforts do not include OCT-A data.

We believe that strategies able to map fundus images to en-face OCT-A can create precise vascular vessel segmentation with less effort.

In this dataset, called UTHealth - Fundus and Synthetic OCT-A Dataset (UT-FSOCTA), we include fundus images and en-face OCT-A images for 112 subjects. The two modalities have been manually aligned to allow for training of medical imaging machine learning pipelines. This dataset is accompanied by a manuscript that describes an approach to generate fundus vessel segmentations using OCT-A for training (Coronado et al., 2022). We refer to this approach as "Synthetic OCT-A".

Fundus Imaging

We include 45 degree macula centered fundus images that cover both macula and optic disc. All images were acquired using a OptoVue iVue fundus camera without pupil dilation.

The full images are available at the fov45/fundus directory. In addition, we extracted the FOVs corresponding to the en-face OCT-A images collected in cropped/fundus/disc and cropped/fundus/macula.

Enface OCT-A

We include the en-face OCT-A images of the superficial capillary plexus. All images were acquired using an OptoVue Avanti OCT camera with OCT-A reconstruction software (AngioVue). Low quality images with errors in the retina layer segmentations were not included.

En-face OCTA images are located in cropped/octa/disc and cropped/octa/macula. In addition, we include a denoised version of these images where only vessels are included. This has been performed automatically using the ROSE algorithm (Ma et al. 2021). These can be found in cropped/GT_OCT_net/noThresh and cropped/GT_OCT_net/Thresh, the former contains the probabilities of the ROSE algorithm the latter a binary map.

Synthetic OCT-A

We train a custom conditional generative adversarial network (cGAN) to map a fundus image to an en face OCT-A image. Our model consists of a generator synthesizing en face OCT-A images from corresponding areas in fundus photographs and a discriminator judging the resemblance of the synthesized images to the real en face OCT-A samples. This allows us to avoid the use of manual vessel segmentation maps altogether.

The full images are available at the fov45/synthetic_octa directory. Then, we extracted the FOVs corresponding to the en-face OCT-A images collected in cropped/synthetic_octa/disc and cropped/synthetic_octa/macula. In addition, we performed the same denoising ROSE algorithm (Ma et al. 2021) used for the original enface OCT-A images, the results are available in cropped/denoised_synthetic_octa/noThresh and cropped/denoised_synthetic_octa/Thresh, the former contains the probabilities of the ROSE algorithm the latter a binary map.

Other Fundus Vessel Segmentations Included

In this dataset, we have also included the output of two recent vessel segmentation algorithms trained on external datasets with manual vessel segmentations. SA-Unet (Li et. al, 2020) and IterNet (Guo et. al, 2021).

• SA-Unet. The full images are available at the fov45/SA_Unet directory. Then, we extracted the FOVs corresponding to the en-face OCT-A images collected in cropped/SA_Unet/disc and cropped/SA_Unet/macula.

• IterNet. The full images are available at the fov45/Iternet directory. Then, we extracted the FOVs corresponding to the en-face OCT-A images collected in cropped/Iternet/disc and cropped/Iternet/macula.

Train/Validation/Test Replication

In order to replicate or compare your model to the results of our paper, we report below the data split used.

• Training subjects IDs: 1 - 25

• Validation subjects IDs: 26 - 30

• Testing subjects IDs: 31 - 112

Data Acquisition

This dataset was acquired at the Texas Medical Center - Memorial Hermann Hospital in accordance with the guidelines from the Helsinki Declaration and it was approved by the UTHealth IRB with protocol HSC-MS-19-0352.

User Agreement

The UT-FSOCTA dataset is free to use for non-commercial scientific research only. In case of any publication the following paper needs to be cited

Coronado I, Pachade S, Trucco E, Abdelkhaleq R, Yan J, Salazar-Marioni S, Jagolino-Cole A, Bahrainian M, Channa R, Sheth SA, Giancardo L. Synthetic OCT-A blood vessel maps using fundus images and generative adversarial networks. Sci Rep 2023;13:15325. https://doi.org/10.1038/s41598-023-42062-9.

Funding

This work is supported by the Translational Research Institute for Space Health through NASA Cooperative Agreement NNX16AO69A.

Research Team and Acknowledgements

Here are the people behind this data acquisition effort:

Ivan Coronado, Samiksha Pachade, Rania Abdelkhaleq, Juntao Yan, Sergio Salazar-Marioni, Amanda Jagolino, Mozhdeh Bahrainian, Roomasa Channa, Sunil Sheth, Luca Giancardo

We would also like to acknowledge for their support: the Institute for Stroke and Cerebrovascular Diseases at UTHealth, the VAMPIRE team at University of Dundee, UK and Memorial Hermann Hospital System.

References

Coronado I, Pachade S, Trucco E, Abdelkhaleq R, Yan J, Salazar-Marioni S, Jagolino-Cole A, Bahrainian M, Channa R, Sheth SA, Giancardo L. Synthetic OCT-A blood vessel maps using fundus images and generative adversarial networks. Sci Rep 2023;13:15325. https://doi.org/10.1038/s41598-023-42062-9.


C. Guo, M. Szemenyei, Y. Yi, W. Wang, B. Chen, and C. Fan, "SA-UNet: Spatial Attention U-Net for Retinal Vessel Segmentation," in 2020 25th International Conference on Pattern Recognition (ICPR), Jan. 2021, pp. 1236–1242. doi: 10.1109/ICPR48806.2021.9413346.

L. Li, M. Verma, Y. Nakashima, H. Nagahara, and R. Kawasaki, "IterNet: Retinal Image Segmentation Utilizing Structural Redundancy in Vessel Networks," 2020 IEEE Winter Conf. Appl. Comput. Vis. WACV, 2020, doi: 10.1109/WACV45572.2020.9093621.

Y. Ma et al., "ROSE: A Retinal OCT-Angiography Vessel Segmentation Dataset and New Model," IEEE Trans. Med. Imaging, vol. 40, no. 3, pp. 928–939, Mar. 2021, doi: 10.1109/TMI.2020.3042802.

We release a human-centric synthetic data generator PeopleSansPeople which contains simulation-ready 3D human assets, a parameterized lighting and camera system, and generates 2D and 3D bounding box, instance and semantic segmentation, and COCO pose labels. Using PeopleSansPeople, we performed benchmark synthetic data training using a Detectron2 Keypoint R-CNN variant [1]. We found that pre-training a network using synthetic data and fine-tuning on target real-world data (few-shot transfer to limited subsets of COCO-person train [2]) resulted in a keypoint AP of 60.37±0.48 (COCO test-dev2017) outperforming models trained with the same real data alone (keypoint AP of 55.80) and pre-trained with ImageNet (keypoint AP of 57.50). This freely-available data generator should enable a wide range of research into the emerging field of simulation to real transfer learning in the critical area of human-centric computer vision.

This synthetic dataset includes the number of visits to a cardiology service associated to a private health insurance per policyholder (column "Visits") in each year of the period 2019-2022. Each row represent a policyholder, and the dataset includes their sex (column "Sex", 0 means female and 1 means male), age group (column "Age", 0 means under 60 and 1 means over 60 years old) and year (column "Year", within the period 2019-2022). As policyholders with no visits in the period were not included in the original dataset, the zeros were synthetically generated by means of a zero-truncated Poisson model, accounting for age category, sex and year. The code used to generate these zeros can be found in https://github.com/dmorinya/BSTS_HealthInsurance (script poisson_cross_sectional_Cardio.R), although the original dataset is not available due to privacy issues. The policy numbers in this dataset (column "Policy") are randomly generated (although the same policy number corresponds to the same policyholder).

The Insider Threat Test Dataset is a collection of synthetic insider threat test datasets that provide both background and malicious actor synthetic data.The CERT Division, in partnership with ExactData, LLC, and under sponsorship from DARPA I2O, generated a collection of synthetic insider threat test datasets. These datasets provide both synthetic background data and data from synthetic malicious actors.For more background on this data, please see the paper, Bridging the Gap: A Pragmatic Approach to Generating Insider Threat Data.Datasets are organized according to the data generator release that created them. Most releases include multiple datasets (e.g., r3.1 and r3.2). Generally, later releases include a superset of the data generation functionality of earlier releases. Each dataset file contains a readme file that provides detailed notes about the features of that release.The answer key file answers.tar.bz2 contains the details of the malicious activity included in each dataset, including descriptions of the scenarios enacted and the identifiers of the synthetic users involved.

The synthetic event log described in this document consists of 25,000 traces, generated using the process model outlined in Geyer et al. (2024) [1] and the DALG tool [2]. This event log simulates the treatment process of malignant melanoma patients, adhering to clinical guidelines. Each trace in the log represents a unique patient journey through various stages of melanoma treatment, providing detailed insights into decision points, treatments, and outcomes.

The DALG tool [2] was employed to generate this data-aware event log, ensuring realistic data distribution and variability.

DALG: https://github.com/DavidJilg/DALG

[1] Geyer, T., Grüger, J., & Kuhn, M. (2024). Clinical Guideline-based Model for the Treatment of Malignant Melanoma (Data Petri Net) (1.0). Zenodo. https://doi.org/10.5281/zenodo.10785431

[2] Jilg, D., Grüger, J., Geyer, T., Bergmann, R.: DALG: the data aware event log generator. In: BPM 2023 - Demos & Resources. CEUR Workshop Proceedings, vol. 3469, pp. 142–146. CEUR-WS.org (2023)

Comparison of observed SRTR and synthetically generated candidate populations.

We create a synthetic administrative dataset to be used in the development of the R package for calculating quality indicators for administrative data (see: https://github.com/sook-tusk/qualadmin) that mimic the properties of a real administrative dataset according to specifications by the ONS. Taking over 1 million records from a synthetic 1991 UK census dataset, we deleted records, moved records to a different geography and duplicated records to a different geography according to pre-specified proportions for each broad ethnic group (White, Non-white) and gender (males, females). The final size of the synthetic administrative data was 1033664 individuals.

National Statistical Institutes (NSIs) are directing resources into advancing the use of administrative data in official statistics systems. This is a top priority for the UK Office for National Statistics (ONS) as they are undergoing transformations in their statistical systems to make more use of administrative data for future censuses and population statistics. Administrative data are defined as secondary data sources since they are produced by other agencies as a result of an event or a transaction relating to administrative procedures of organisations, public administrations and government agencies. Nevertheless, they have the potential to become important data sources for the production of official statistics by significantly reducing the cost and burden of response and improving the efficiency of such systems. Embedding administrative data in statistical systems is not without costs and it is vital to understand where potential errors may arise. The Total Administrative Data Error Framework sets out all possible sources of error when using administrative data as statistical data, depending on whether it is a single data source or integrated with other data sources such as survey data. For a single administrative data, one of the main sources of error is coverage and representation to the target population of interest. This is particularly relevant when administrative data is delivered over time, such as tax data for maintaining the Business Register. For sub-project 1 of this research project, we develop quality indicators that allow the statistical agency to assess if the administrative data is representative to the target population and which sub-groups may be missing or over-covered. This is essential for producing unbiased estimates from administrative data. Another priority at statistical agencies is to produce a statistical register for population characteristic estimates, such as employment statistics, from multiple sources of administrative and survey data. Using administrative data to build a spine, survey data can be integrated using record linkage and statistical matching approaches on a set of common matching variables. This will be the topic for sub-project 2, which will be split into several topics of research. The first topic is whether adding statistical predictions and correlation structures improves the linkage and data integration. The second topic is to research a mass imputation framework for imputing missing target variables in the statistical register where the missing data may be due to multiple underlying mechanisms. Therefore, the third topic will aim to improve the mass imputation framework to mitigate against possible measurement errors, for example by adding benchmarks and other constraints into the approaches. On completion of a statistical register, estimates for key target variables at local areas can easily be aggregated. However, it is essential to also measure the precision of these estimates through mean square errors and this will be the fourth topic of the sub-project. Finally, this new way of producing official statistics is compared to the more common method of incorporating administrative data through survey weights and model-based estimation approaches. In other words, we evaluate whether it is better 'to weight' or 'to impute' for population characteristic estimates - a key question under investigation by survey statisticians in the last decade.