Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
4,612
datasets available to search
ShareScore release 0.9.0
Dataset results
4,612 results for “labelled”
CRLC: Cross-resolution national-scale land-cover mapping based on noisy label learning: a case study of China
<p><strong>CRLC are the 10-meter resolution land cover maps for China in 2020 achieved by a deep classification network.</strong></p> <p><strong>The maps include eight land cover classes:</strong></p> <p>1: Cropland</p> <p>2: Forest</p> <p>3: Grass/shrubland</p> <p>5: Wetland</p> <p>6: Water bodies</p> <p>8: Impervious</p> <p>9: Bareland</p> <p>10: Snow/ice</p> <p> </p> <p><strong>Reference:</strong></p> <pre>@article{liu2023cross, title={Cross-resolution national-scale land-cover mapping based on noisy label learning: A case study of China}, author={Liu, Yinhe and Zhong, Yanfei and Ma, Ailong and Zhao, Ji and Zhang, Liangpei}, journal={International Journal of Applied Earth Observation and Geoinformation}, volume={118}, pages={103265}, year={2023}, publisher={Elsevier} }</pre>
MICCAI FLARE22 Challenge Dataset (50 Labeled Abdomen CT Scans)
<p>This dataset was used as the labeled training set in MICCAI FLARE 2022 Challenge https://flare22.grand-challenge.org/.</p> <p>The CT images and pancreas annotations are from http://medicaldecathlon.com/</p> <p>The other organ annotations are from AbdomenCT-1K (research purpose only).</p> <p>If this dataset is useful in your research, please give credit to the following two papers:</p> <p>A large annotated medical image dataset for the development and evaluation of segmentation algorithms</p> <p>https://arxiv.org/abs/1902.09063</p> <p>AbdomenCT-1K: Is Abdominal Organ Segmentation a Solved Problem?</p> <p>https://ieeexplore.ieee.org/document/9497733</p>
Sentinel-2 Wind Turbine Images with Spin and Spin Quality Labels
<p>Reproduction Data for training a classifier on the Sentinel-2 Band Offset for motion detection.</p>
Dataset for "Generalizing property prediction of ionic liquids from limited labeled data: a one-stop framework empowered by transfer learning"
<p>Dataset for "Generalizing property prediction of ionic liquids from limited labeled data: a one-stop framework empowered by transfer learning", codes can be found <a href="https://github.com/GuzhongChen/ILTransR">here</a>.</p>
Background-Foreground-Segmentation Labels for "Self-improving Semantic Perception on a Construction Robot"
<p>Background-foreground-segmentation labels the paper "Self-improving Semantic Perception on a Construction Robot" of CoRL 2021</p>
Figure1 in Lost and found in Ireland; how a data label resulted in a postal delivery to Metriocnemus (Inermipupa) carmencitabertarum (Orthocladiinae)
Figure1. The internally labelled vial that was lost in the post
Label-free adaptive optics single-molecule localization microscopy for whole zebrafish
<p>The specimen-induced aberration has been a major factor limiting the imaging depth of single-molecule localization microscopy (SMLM). Here, we report the application of label-free wavefront sensing adaptive optics to SMLM for deep-tissue super-resolution imaging. The proposed system measures complex tissue aberrations from intrinsic reflectance rather than fluorescence emission and physically corrects the wavefront distortion more than three-fold stronger than the previous limit. This enables us to resolve sub-diffraction morphologies of cilia and oligodendrocytes in whole zebrafish as well as dendritic spines in thick mouse brain tissues at the depth of up to 102 μm with localization number enhancement by up to 37 times and localization precision comparable to aberration-free samples. The proposed approach can expand the application range of SMLM to whole zebrafish that cause the loss of localization points owing to severe tissue aberrations.</p>
Psilocybin therapy for females with anorexia nervosa: A phase 1, open label, feasibility study
<p class="MsoNoSpacing">Anorexia nervosa (AN) is a deadly illness with no proven treatments to reverse core symptoms and no FDA-approved medications. Novel treatments are urgently needed to improve clinical outcomes. In this open-label feasibility study, 10 adult female participants (<em>m</em> BMI 19.7 kg/m<sup>2</sup>; SD 3.7) who met DSM 5 criteria for AN or pAN (partial remission) were recruited to a study conducted at an academic clinical research institute. Participants received a single 25mg dose of synthetic psilocybin combined with psychological support. The primary aim was to assess safety, tolerability, and feasibility at post-treatment via incidences and occurrences of adverse events (AEs) and clinically significant changes in: ECG, laboratory tests, vital signs and suicidality. There were no clinically significant changes in ECG, vital signs, or suicidality. Two participants developed asymptomatic hypoglycemia at post-treatment, which resolved within 24 hours. There were no other clinically significant changes in laboratory values. All AEs were mild and transient in nature. Participants' qualitative perceptions suggest that the treatment was acceptable for most participants. <span>R</span>esults suggest that psilocybin therapy is safe, tolerable, and acceptable for female AN, a promising finding given physiological dangers and problems with treatment engagement.</p> <p class="MsoNoSpacing">ClinicalTrials.gov Identifier: NCT04661514</p>
scNCL transfers labels from scRNA-seq to scATAC-seq data with neighborhood contrastive regularization
<p>Data used in our manuscript.</p> <p>'scNCL_data' folder contains gene expression data (with protein) and gene activitiy data (with protein).</p> <p>'GLUE_data' folder contains gene expression data and raw chromatin accessibility data. </p> <p>'HFA-resampling' folder contains multiple independent sampling of HFA-subset dataset.</p> <p>'Accuray on HFA-subsets.xlsx' file contains accuray results of all compared methods on HFA-subsets datasets.</p>
Structure preserving adversarial generation of labeled training samples for single cell segmentation
<p>Supplementary package (dataset, generated data and code) for the article: "<strong>Structure preserving adversarial generation of labeled training samples for single cell segmentation</strong>". (Cell Reports Methods, 2023 - under submission)</p> <p>Content of the package:</p> <ul> <li>datasets.zip: the images and annotations for the fallopian tube and salivary gland dataset.</li> <li>mask_quality_experiment.zip: the generated masks using SIMCEP and the generated microscopy images using pix2pix using the SIMCEP masks.</li> <li>pix2pix_models.zip: the models for fold 0 for both datasets. It can be used to synthesize new images.</li> <li>StyleGAN2-ada_models.zip: the models for synthesizing new masks.</li> <li>code.zip: a github repository (named .git) for executing the pipeline.</li> </ul>
Antimagic Labelings of All Trees of Order 3 to 25
<p>This archive contains antimagic labelings for all trees of order 3 to 25 found by evolutionary search as described in the paper: Branson, L. and Sutton, A. M., <em>Finding Antimagic Labelings of Trees by Evolutionary Search</em>, 2023. https://doi.org/10.1145/3594805.3607133</p> <p>The archive contains 1679 compressed sub-archives, each containing 100000 data files, except the last, which contains 79144 data files. The total number of data files is 167879144 (cf. the total number of unlabeled trees of order 3-25 in entry A000055 of the OEIS https://oeis.org/A000055). Each data file corresponds to a unique tree and is named <strong>tree-<n>-<id>.viz</strong> where <strong><n></strong> is the order (number of vertices) of the tree and <strong><id></strong> is a unique identifier for the tree of that order. Each data file is a text file written in the DOT graph description language that specifies the structure of the tree along with an edge labeling and vertex labeling that is antimagic.</p>
NPM3D dataset with instance label. Dataset used in paper "A Review of Panoptic Segmentation for Mobile Mapping Point Clouds"
<p>NPM3D is a public benchmark for point cloud semantic segmentation, with 10 classes including: ground, building, pole (road sign and traffic light), bollard, trash can, barrier, pedestrian, car, natural (vegetation) and unclassified. Results are evaluated only w.r.t. 9 classes, disregarding the "unclassified" label. The data has been captured with a mapping-grade mobile laser scanning system in different cities in France. There are 4 regions designated for training, all captured in Paris and Lille; and 3 regions for testing, captured in Dijon and Ajaccio. The standard 10-class version described above has actually been derived from a more fine-grained version of the dataset by keeping only the most frequent labels. The original annotations feature 50 different semantic classes (most of which are very rare), and also individual object instance labels for the training regions. For panoptic segmentation, a new version has been generated that still uses the 10 semantic category labels listed above, but also includes instance labels. The classes ground, building and barrier are considered "stuff" and are not separated into instances. As no instance labels are available for the 3 test regions, our version for panoptic (or pure instance) segmentation only contains 4 different regions from Paris and Lille. Instead of a fixed training/test split all experiments therefore use 4-fold cross-validation.</p>
Cardiac Cine MRI images and labels of Sus scrofa at 7T
<p>Cardiac Cine MRI images and labels of Sus scrofa at 7T. See <a href="https://github.com/chfc-cmi/cmr-cine-sscrofa/tree/master">repository on GitHub</a>.</p>
Evaluation of local movement of Bemisia tabaci MEAM1 and tomato severe rugose virus transmission to tomato plants using labeled and unlabeled whiteflies
<p>The data presents the effects of two types of labels on whiteflies' behavior. Besides, it shows the flight distances at close range and ToSRV transmission.</p>
Remote Sensing and Geomorphological Covariates and Labels of main Peruvian Glaciers
<p>Abstract: The Peruvian Glacier Classification Dataset is a comprehensive collection designed to facilitate research and analysis in the field of glaciology and remote sensing. Comprising a total of 9 rasterStacks, each representing distinct glacier regions within Peru, the dataset offers a rich set of 25 morphological and Landsat 8 derived covariates for in-depth study. These covariates have been meticulously selected to capture a wide range of features and characteristics associated with glaciers and their surrounding environments. Moreover, the dataset includes true labels indicating the classification of each pixel into Glacier and Non-Glacier classes, as determined by the National Inventory of 2017.</p> <p>Description: The Peruvian Glacier Classification Dataset presents a valuable resource for researchers, scientists, and practitioners interested in glacial dynamics, environmental monitoring, and geospatial analysis. Comprising 9 main glacier regions within Peru (Blanca, Central, Huallanca, Huayhuasha, Huaytapallana, Raura,Urubamba, Vilcabamba,Vilcanota), the dataset provides an extensive collection of covariates derived from both morphological features and Landsat 8 satellite imagery. These covariates have been processed and curated to offer comprehensive insights into the intricate nature of glaciers and their surroundings.</p> <p>Key Features:</p> <ol> <li> <p>Morphological Covariates: The dataset encompasses a diverse array of morphological features extracted from high-resolution elevation data calculated with SAGA GIS. </p> </li> <li> <p>Landsat 8 Derived Covariates: Landsat Covariates corresponding to the period 2017–2018 obtained from Landsat Collection 2 Level 2 and Tier 1 surface reflectance (SR) products (Vermote et al., 2016) available online: https://www.usgs.gov/landsat-missions/landsat-collection-2-surface-reflectance (accessed on 1 July 2023). procesing with GoogleEarth Engine study. https://code.earthengine.google.com/fea31b7f2a3fdbc1065644616819134a. This includes the following covariates: BLUE, GREEN, RED, NIR, SWIR1, SWIR2, NDFI, ndsi, ndvi, NDWI , NDWIns, NDSInw, nbr2, VNSIR, NDMI, TCG .</p> </li> <li> <p>True Labels: Ground truth information is provided through accurate classification labels for each pixel, classifying it as either belonging to the Glacier or Non-Glacier class. These labels are based on the authoritative National Inventory of 2017, enhancing the reliability and usability of the dataset.</p> </li> </ol>
QM9-XAS database of 56k QM9 small organic molecules labeled with TDDFT X-ray absorption spectra
<p>Database for training graph neural network (GNN) models in <strong>Integrating Explainability into Graph Neural Network Models for the Prediction of X-ray Absorption Spectra, </strong>by Amir Kotobi, Kanishka Singh, Daniel Höche, Sadia Bari, Robert H.Meißner, and Annika Bande.</p> <p><strong>Included:</strong></p> <ul> <li>qm9_Cedge_xas_56k.npz: the TDDFT XAS spectra of 56k structures from the QM9 dataset, were employed to label the graph dataset. The dataset contains two pairs of key/value entries: <strong>spec_stk</strong>, which represents a 2D array containing energies and oscillator strengths of XAS spectra, and <strong>id</strong>, which consists of the indices of QM9 structures. This data was used to create the QM9-XAS graph dataset.</li> <li>qm9xas_orca_output.zip: the raw ORCA output of TDDFT calculations for the 56k QM9-XAS dataset consists of excitation energies, densities, molecular orbitals, and other relevant information. This unprocessed output serves as a source to derive ground truth data for explaining the predictions made by GNNs.</li> <li>qm9xas_spec_train_val.pt: processed graph train/validation dataset of 50k QM9 structures. It is used as input to GNN models for training and validation.</li> <li>qm9xas_spec_test.pt: processed graph test dataset of 6k QM9 structures. It is used to test the performance of trained GNN models.</li> </ul> <p><strong>Notes on the datasets:</strong></p> <ul> <li>The QM9-XAS dataset was created using ORCA electronic structure package [Neese, F., WIREs Computational Molecular Science 2012, 2, 73–78] to calculate carbon K-edge XAS spectra with the time-dependent density functional theory (TDDFT) method [Petersilka, M.; Gossmann, U. J.; Gross, E. K. U., Phys. Rev. Lett. 1996, 76, 1212–1215]</li> <li>The molecular structures of QM9-XAS datasets were sourced from the QM9 database [R. Ramakrishnan, P. O. Dral, M. Rupp, and O. A. Von Lilienfeld, <em>Sci. Data</em> 1, 1 (2014)].</li> </ul> <p><strong>Funding:</strong></p> <p>This<strong> </strong>research was funded by HIDA Trainee Network program, HAICU, Helmholtz AI-4-XAS, DASHH and HEIBRiDS graduate schools. For theoretical calculations and model training, computational resources at DESY and JFZ were used. </p>
Labels for land cover classification for Sentinel 2 images in livestock farms
<p>Land cover data labeling for 2022 Sentinel 2 images in livestock farms, mostly in different agroclimatic regions of Spain. The data has a resolution of 10m and the label values correspond to:</p> <ul> <li>0: Unproductive.</li> <li>1: Woodland.</li> <li>2: Pasture.</li> </ul> <p>Dataset produced within the <a href="https://ai4copernicus-project.eu/">AI4Copernicus</a> European H2020 project.</p>
Field testing a 13C labeling method in an East African ant-plant
<p>Tree carbon allocation is a dynamic process that depends on the tree's environment, but we know relatively little about how biotic interactions influece these dynamics. In central Kenya, the loss of vertebrate herbivores and the savanna's invasion by the ant <em>Pheidole megacephala</em> are disrupting mutualisms between the founational tree <em>Acacia drepanolobium</em> and its native ant defenders. Here we piloted a <sup>13</sup>Carbon (C) pulse-labeling mathod to investigate the influece of these biotic interactions on C allocation strategies by adult trees in situ. Trees withstood experimental conditions and took up sufficient labeled <sup>13</sup>CO<sub>2 </sub>for <sup>13</sup>C to be detected in various C sinks, including ant mutualists. The <sup>13</sup>C in ants collected shortly after labeling suggested that trees exposed to herbivores allocated relatively more newly assimilated C to native ant defenders. Our results demonstrate the viability of the pulse-labeling method and suggest the C allocation to ant partners depends on the biotic context of the tree, but further investigation with replication is needed to characterize such differences in relation to invasion and herbivore loss.</p>
Data compilation for "Label-free Imaging of Catalytic H2O2 Decomposition on Single Colloidal Pt Nanoparticles using Nanofluidic Scattering Microscopy"
<p>Data compilation for "Label-free Imaging of Catalytic H2O2 Decomposition on Single Colloidal Pt Nanoparticles using Nanofluidic Scattering Microscopy"</p> <p>This data set contains raw and evaluated data for the publication mentioned above and is structured into three parts which are packed as .zip files. The file labeled with MAIN 1 contains data for the figures 1 to 3 and MAIN 2 for figures 4 to 6 of the main manuscript. The file labeled with SI contains all data related to the supplementary information.</p> <p>The data itself has three main types. There are .tif images which contain the raw darkfield microscopy images, usually as image stack of several single pictures. Some of these .tif images have been processed according to the description in the publication and single frames exported as .png images.<br> The .xlsx files contain the evaluated data that has been extracted from the microscopic images and was used to draw the corresponding graphs in the publication. The individual tables and the respective columns/rows have a basic labeling identifying their contents.<br> In addition, the data set also contains SEM/TEM images that show mainly the platinum particles used in the experiments. There is also one Comsol simulation for the flow in a nanochannel that is being filled with a bubble, which was created with Comsol 5.5.</p>
Micro-CT tomographic data set of 38 mummy labels from the BNU in Strasbourg (2/2)
<p><strong>Summary</strong></p> <p>This submission contains a tomographic dataset of 38 mummy labels from the BNU in Strasbourg used to perceive the anatomical identification possibilities of the woods used for mummy labels and to carry out ring width measurements. The data will be made available as part of [Blondel et al., 2024].</p> <p><strong>Apparatus</strong></p> <p>The dataset is acquired using the EasyTom 150/160 X-ray tomograph (RX Solutions). This tomograph is equipped with a sealed X-ray generator with a compact tube and an interchangeable-plane sensor fitted with a CsI scintillator. The CT scanner parameters for the session carried out on the mummy labels were set at 90 Kv with an intensity of 195 mA for an acquisition resolution varying between 11 and 42 µm with 2016 projections (that is about 20 images on average per projection) with a frame rate of 12,5 and a temperature of 28°C. Each image was then reconstructed by filtered retroprojection using the XAct software (RX Solutions).</p> <p><strong>Information on placing mummy labels in the tomograph</strong></p> <p>The installation of the mummy labels was the same for all the different labels, some of which varied in size. They were attached to a plastic clamping vice-type support covered in expanded foam to prevent the labels from being marked during clamping, before being placed on the tomograph's rotating platform.</p> <p><strong>Issues relating to the data collected</strong></p> <p>The data collected for this study were carried out to perceive the possibilities of anatomical identification from tomographic images in the transverse plane. The tangential and radial planes were not of sufficiently high resolution due to the dimensions of the mummy labels, see details in [Blondel et al., 2024]. The other objective was to use tomographic imagery to facilitate the acquisition of ring widths in the transverse plane of mummy labels. The mummy labels were not tomographed in their entirety. Only the central part, a few centimetres high, was tomographed to maximise resolution. The number of projections and the resolution per label are specified in table form in [Blondel et al., 2024], as they vary according to the width and thickness of the mummy labels. All raw tomography image data (i.e. without corrections) are available in .tif format. The post-processing steps are described in the methodology of [Blondel et al., 2024].</p> <p><strong>List of Contents</strong></p> <p>The content of the submission is divided into 38 data sets corresponding to the 38 mummy labels. Each set is labelled with the inventory number of the BNU mummy label and its resolution. Each set contains:<br>- all the images of the transverse plane in .tif format, the number of projections of which varies from one label to another depending on the resolution of the acquisitions, see details in [Blondel et al., 2024].<br>- The .xls file containing a summary of the scanner metadata for each of the mummy labels.<br>- The three images processed in the transverse plane for each label, including those used to measure ring width for the 7 labels for which ring width measurement was possible, as presented in [Blondel et al., 2024].<br>- Colour photographs of the front and back of each tomographed mummy label including those on which ring width measurements were taken on their surface, unless otherwise stated<a title="" href="#_ftn1" name="_ftnref1">[1]</a>. All these photographs are marked: Coll._et_photogr._BNU_Strasbourg_OpenLicence, accompanied by the inventory number.</p> <p><strong>Acknowledgments</strong></p> <p>We would also like to thank engineers Damien Favier and Antoine Egele from the Charles Sadron Institute for their work on the tomographic acquisitions carried out on the 38 mummy labels.</p> <div><br> <div> <p><a title="" href="#_ftnref1" name="_ftn1">[1]</a> The photographs of the front and back of mummy label HO255 are not available, as they are currently being studied.</p> </div> </div>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.