Find research datasets worth reusing

opencc-by-4.0Apr 2024View details →

Fetal exposure to the Ukraine famine of 1932-1933 and adult Type 2 Diabetes Mellitus (Public data and analytical code)

Abstract The short-term impact of famines on death and disease is well documented but it is difficult to estimate their potential long-term impact. We used the setting of the man-made Ukrainian Holodomor famine of 1932-1933 to examine the relationship between prenatal famine and adult Type 2 diabetes mellitus (T2DM). This ecological study included 128,225 T2DM cases diagnosed between 2000-2008 among 10,186,016 male and female Ukrainians born between 1930 and 1938. Individuals who were born in the first half-year of 1934, and hence exposed in early gestation to the mid-1933 peak famine period, had a larger than two-fold likelihood of T2DM (OR 2.21; 95% CI 2.00-2.45) compared to unexposed controls. There was a dose-response relationship between severity of famine exposure and adult T2DM risk comparing individuals born in regions with severe, very severe, and extreme famine to births in the no-famine region.   Description of the data and analytical code In exploratory analyses we first examined whether the odds for T2DM were elevated for any month of birth in the period January 1930 to December 1938 in any of the four regions of varying famine intensity. This was achieved by comparing, within each region, the T2DM odds for births in any month and year of birth relative to the T2DM odds for births in the same month combining all other years of birth. The analysis served to identify potential relations of famine with specific months and years of birth, controlling for month of birth effects. We observed increased T2DM odds ratios for births between January and June 1934 in famine-exposed oblasts, with smaller increases for births in 1935 and 1936 in these months. Our findings suggested that in multivariate modelling statistical control for month of birth effects could be accomplished by adjusting for the January-June period. Our findings are presented in the data file '01 Odds Ratio for T2DM Over Time' and show the odds ratios (ORs) for Type 2 Diabetes Mellitus (T2DM) comparing the region-specific T2DM odds for each birth year and month relative to births in the same months but combining all other years of birth. The R syntax file '01 Odds of T2DM Over Time Figure' provides the code necessary to reproduce the figure.   For confirmatory analyses we employed a Difference-in-Differences approach to quantify associations between prenatal exposure to famine and T2DM, taking into account year of birth, half-year of birth (Jan-Jun vs Jul-Dec), region, and their interactions. This analysis was conducted initially for each gender separately and then for both genders combined, adjusting for We carried out sensitivity analyses to assess potential changes in T2DM odds arising from the use of pre-famine births vs post-famine births as controls.  Our findings are presented in the data file '02 Ukraine Famine 1932-33 Main Data'. Information on the number of T2DM cases by gender, region of residence, and year and month of birth 1930-1938 in Ukraine was collected by the national Ukraine Diabetes Register (Komisarenko Institute of Endocrinology and Metabolism, Kyiv) between 2000-2008. The number of births in the same subgroups, representing the populations at risk for T2DM, was estimated by demographic population reconstruction methods as reported in the publication. We classified the birth counts by year of birth, the semi-annual birth period (January-June vs. July-December), region of birth, and gender. The SPSS syntax file titled '02 Ukraine Famine 1932-33 Main Analysis' provides the code to replicate our main findings as presented in the publication.   In a separate analysis we visualized by a meta-regression approach the relation between famine intensity at the oblast level in 1933 and the odds for adult T2DM.  The data required for the replication of our findings are included in the file '03 Odds Ratio for T2DM and Famine Intensity at Oblast Level'. The R syntax file titled '03 Ukraine Famine 1932-33 Meta-regression' provides details on conducting the meta-regression using the R package ‘metafor’.   Funding Ukraine State complex program Diabetes Mellitus, project number 0106U000844 (M.K.). Holodomor Research and Education Consortium in Canada (L.H.L., O.W.). NIDI-NIAS Fellowship of the Royal Netherlands Academy of Sciences (L.H.L.). National Institute of Aging R01 AG028593 (L.H.L.). National Institute of Aging R01 AG06687 (L.H.L.).   Sharing/Access information Data sharing and use are unrestricted with acknowledgement of the original publication and listing of the funding sources as per the above. Researchers are encouraged to contact the Principal Investigators (PIs) for consultations on data structure and use as needed (L.H. Lumey, <a href="mailto:lumey@columbia.edu">lumey@columbia.edu</a>; Oleh Wolowyna, <a href="mailto:olehw@aol.com">olehw@aol.com</a>).

opencc-by-4.0Nov 2024View details →

NatalIA: PBF-US1 (Phantom Blind-sweeps for Fetal Ultrasound Scanning)

NatalIA PBF-US1 is dataset designed to support the development of AI-based tools for detecting relevant fetal planes in ultrasound videos captured by non-trained personnel, such as midwives or nurses.

openother-openJul 2021View details →

A Spatio-temporal Atlas of the Developing Fetal Brain with Spina Bifida Aperta

This version contains two zipped folders. <ol> <li><a href="https://zenodo.org/api/files/c84cc018-9d2d-4adb-9558-9ab96649c922/spina_bifida_atlas.zip">spina_bifida_atlas.zip</a> contains a copy of our spina bifida aperta fetal brain atlas. This folder is available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication), as indicated in the LICENSE file in this folder. This is the same version of the atlas as the one available on synapse (<a href="https://www.synapse.org/#!Synapse:syn25887675/wiki/611424">https://www.synapse.org/#!Synapse:syn25887675/wiki/611424</a>, DOI: 10.7303/syn25887675).</li> <li><a href="https://zenodo.org/api/files/c84cc018-9d2d-4adb-9558-9ab96649c922/LucasFidon/spina-bifida-MRI-atlas-0.1.0.zip?versionId=31751dfa-7b17-4c84-89d5-2769d648eed8">LucasFidon/spina-bifida-MRI-atlas-0.1.0.zip</a> is a copy of the code that was used to compute the fetal brain atlas for spina bifida aperta in this repository. This folder is available under BSD-3-Clause license, archived from GitHub, as indicated in the LICENSE file in this folder.</li> </ol> How to cite: If you find the data in this folder useful for your research please cite: L. Fidon, E. Viola, N. Mufti, A. L. David, A. Melbourne, P. Demaerel, S. Ourselin, T. Vercauteren, J. Deprest, M. Aertsen. A Spatio-temporal Atlas of the Developing Fetal Brain with Spina Bifida Aperta, 2021.

opencc-by-4.0Feb 2023View details →

Maternal fetal ultrasound planes from low-resource imaging settings in five African countries

This resource is a dataset of routinely acquired maternal-fetal screening ultrasound images collected in five centers of five countries in Africa (Malawi, Egypt, Uganda, Ghana and Algeria) that is associated to the journal article Sendra-Bacells et al. "Generalisability of fetal ultrasound deep learning models to low-resource imaging settings in five African countries", Scientific Reports. The images correspond to the four most common fetal planes: abdomen, brain, femur and thorax. A CSV file is provided where image filenames are associated to plane types and patient number as well as the partitioning in training and testing splits as used in the associated publication.

opencc-by-4.0Nov 2023View details →

Dataset - FetMRQC: an open-source machine learning framework for multi-centric fetal brain MRI quality control

This dataset contains the data and model used in the paper <blockquote> Thomas Sanchez, Oscar Esteban, Yvan Gomez, Alexandre Pron, Mériam Koob, Vincent Dunet, Nadine Girard, Andras Jakab, Elisenda Eixarch, Guillaume Auzias, and Meritxell Bach Cuadra. "FetMRQC: an open-source machine learning framework for multi-centric fetal brain MRI quality control." <a href="https://arxiv.org/abs/2311.04780">arXiv preprint arXiv:2311.04780</a> (2023). </blockquote> If you found this dataset useful or used it in your research, please cite this reference. This dataset contains manual quality annotations and image quality metrics (IQMs) obtained from 1647 stacks of T2-weighted (T2w) slices of fetal brain magnetic resonance (MR) images collected from 233 subjects at four different institutions Lausanne University Hospital (CHUV) in Switzerland, BCNatal at Hospital Sant Joan de Déu in Barcelona (Spain), University Children's Hospital Zürich (KISPI) in Switzerland and La Timone University Hospital in Marseille, France. The data were acquired on scanners from different vendors (Siemens at CHUV, BCNatal and Marseille, General Electrics at KISPI), MR sequences (Half Fourier Single-shot Turbo spin-Echo –HASTE– for Siemens scanners and Single-Short Fast Spin Echo –SS-FSE– for GE scanners), magnetic field strengths (1.5 T and 3 T), image resolutions, fields of view, repetition times and echo times, with both neurotypical and pathological cases. These data and the derived IQMs were used to train and evaluate models for quality assessment and quality control of fetal brain MR images. The code to reproduce the experiments is available on <a href="https://github.com/Medical-Image-Analysis-Laboratory/fetal_brain_qc">GitHub.</a> Each entry describe the information for a single stack of T2w slices. It contains information regarding which subject it belongs to, its manual quality rating, scanner-related information and 332 IQMs, starting at the `centroid` column in the file. Further description of the data is available in the materials and methods section of the <a href="https://arxiv.org/abs/2311.04780">paper</a>. The model is a 2D nnUNet [1] segmentation network trained on the super-resolution reconstructed data and manual segmentations available as part of the<a href="https://www.synapse.org/#!Synapse:syn25649159/wiki/610007"> Fetal Tissue Annotation Challenge</a> (FeTA). Copyright (c) - All rights reserved. Medical Image Analysis Laboratory - Department of Radiology, Lausanne University Hospital (CHUV) and University of Lausanne (UNIL), Lausanne, Switzerland & CIBM Center for Biomedical Imaging. 2023. [1] Isensee, Fabian, et al. "nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation." Nature methods 18.2 (2021): 203-211.

opencc-by-4.0Nov 2024View details →

Distinct tissue-dependent composition and gene expression of human fetal innate lymphoid cells

Countmatrix and metadata for gene expression in bulk NK cells and CD304+ ILC3s isolated from human fetal liver, lung, intestine, and skin.

Figure 5 in Development of predictive models for determining fetal age-at-length in belugas (Delphinapterus leucas) and their application toward in situ and ex situ population management

Figure 5. Illustration of the linear relationships between fetal age and growth measurements of biparietal diameter (BPD: top graph), thoracic diameter (TD: middle graph) and thoracic circumference (TC: bottom graph) in belugas.

Figure 4 in Development of predictive models for determining fetal age-at-length in belugas (Delphinapterus leucas) and their application toward in situ and ex situ population management

Figure 4. Comparisons of regression curves of TL growth during the first (●) and second half (○) of gestation (top graph) and the early (●), mid (○) and late (▲) pregnancy (bottom graph). The slopes of the regression lines for first half of gestation (F = 63.31, P <0.0001, df1 = 1, df2 = 33) and for early (F = 50.05, P <0.0001, df1 = 1, df2 = 37) and mid pregnancy (F = 135.04, P <0.0001, df1 = 1, df2 = 32) were different than those for the second half of gestation and late pregnancy, respectively. Note that the animals double in length during late pregnancy (315–473 d).

Figure 3 in Development of predictive models for determining fetal age-at-length in belugas (Delphinapterus leucas) and their application toward in situ and ex situ population management

Figure 3. Individual growth rate data from three animals (Animal 1, 2, 3). Regression line slopes during the first two-thirds of pregnancy (top graph) were similar (F = 0.48, P = 0.62, df1 = 2, df2 =18), while regression slopes where different (F = 15.13, P = 0.03, df1 = 2, df2 =3) from the second half to term. Animal 1 (▲) did not have any TL data beyond the first half of gestation so TL length data were used from the farthest in gestation and then again at term. Note that while growth rates were similar during the first two-thirds of pregnancy, fetuses were already different in size when initially detected.

Figure 2 in Development of predictive models for determining fetal age-at-length in belugas (Delphinapterus leucas) and their application toward in situ and ex situ population management

Figure 2. Fetal growth curve comparison illustrating different growth rates resulting in wide range in estimated gestation length as compared to known gestation length determined in this study. Data from Heide-Jørgensen and Teilmann (1994; dotted line) predicts a gestation length of 310 d for a 150 cm calf and similar to our study used a 2nd order polynomial regression to describe their data. Kleinenberg et al. ([1964] 1969: dashed line) developed a curve of the average monthly embryo/fetal growth. They did not provide the curve, only the predicted age at TL, which we then used to fit to a growth curve, which predicts 150 cm calf as 338 d.

Figure 1 in Development of predictive models for determining fetal age-at-length in belugas (Delphinapterus leucas) and their application toward in situ and ex situ population management

Figure 1. Ultrasonographic images of beluga fetuses. All images have yellow caliper lines used to measure dimensions. Biparietal diameter (A, B) at two different stages of gestation show the ovoid shaped skull and echo produced from falx (arrows) located midline between the parietal bones (arrowheads). The thoracic diameter (C) as measured between the yellow caliper marks (arrowheads) on the lateral side of the fetal thorax (d1 = 6.66 cm) at the level of the heart (white arrow) and thoracic circumference (c = 24.04 cm) determined by using the elliptical measurement caliper function to include the dorsal to ventral diameter (d2 = 8.67 cm). The total length of a fetus (D) which is bent in utero, thus requiring the addition of two separate measurements (arrowheads), 1) 8.38 cm from the cranial most aspect of the skull to mid abdomen and 2) 6.91 cm from mid abdomen to distal most portion of the peduncle for a total length of 15.29 cm.

opencc-by-4.0Nov 2019View details →

fetal movement accelerometer data

This dataset contains accelerometer-recorded fetal movement data acquired with healthy volunteers (pregnant women).

opencc-by-4.0Nov 2019View details →

fetal movement acceleration data

This dataset contains accelerometer-recorded fetal movement signals with 16 different pregnant women The dataset contains signals from one accelerometer placed on the abdominal wall of the pregnant women. Note: The file format is .mat, which is recognized by the Matlab software.  the "xxxxxx_bp" files contain the signals of the maternal perception push-button markers (the mother was required to press a push-button every time she felt one fetal movement), and the "xxxxxx_signal" files contain acceleration data from the 3 axes (x, y, and z). Hardware settings:  accelerometer: ADXL355 from ANALOG DEVICES Sampling frequency: 500Hz

opencc-by-4.0Dec 2020View details →

Dataset: Quantitative Evaluation of Enhanced multi-plane clinical fetal diffusion MRI with a crossing-fiber phantom

This dataset provides MRI acquisitions of a customized crossing phantom for fetal brain. It contains: 1) High Resolution acquisitions of 1.5 mm3 isotropic and 61 directions (with b-vectors/b-values) 2) Six low resolution acquisitions of 1x1x4 mm3 and 9-16-25 directions (with respective b-vectors/b-values) 3) A structural T2-w acquisition

dryad40/100

Towards a more informative representation of the fetal-neonatal brain connectome using Variational Autoencoder

Recent advances in functional magnetic resonance imaging (fMRI) have helped elucidate previously inaccessible trajectories of early-life prenatal and neonatal brain development. To date, the interpretation of fetal-neonatal fMRI data has relied on linear analytic models, akin to adult neuroimaging data. However, unlike the adult brain, the fetal and newborn brain develops extraordinarily rapidly, far outpacing any other brain development period across the lifespan. Consequently, conventional linear computational models may not adequately capture these accelerated and complex neurodevelopmental trajectories during this critical period of brain development along the prenatal-neonatal continuum. To obtain a nuanced understanding of fetal-neonatal brain development, including non-linear growth, for the first time, we developed quantitative, systems-wide representations of brain activity in a large sample (>500) of fetuses, preterm, and full-term neonates using an unsupervised deep generative model called Variational Autoencoder (VAE), a model previously shown to be superior to linear models in representing complex resting state data in healthy adults. Here, we demonstrated that non-linear brain features, i.e., latent variables, derived with the VAE pretrained on rsfMRI of human adults, carried important individual neural signatures, leading to improved representation of prenatal-neonatal brain maturational patterns and more accurate and stable age prediction in the neonate cohort compared to linear models. Using the VAE decoder, we also revealed distinct functional brain networks spanning the sensory and default mode networks. Using the VAE, we are able to reliably capture and quantify complex, non-linear fetal-neonatal functional neural connectivity. This will lay the critical foundation for detailed mapping of healthy and aberrant functional brain signatures that have their origins in fetal life.

opencc-zeroMay 2023View details →