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.
328
datasets available to search
ShareScore release 0.9.0
Dataset results
328 results for “segmentation dataset”
Datasets of "Influence of contrast and texture based image modifications on the performance and attention shift of U-Net models for brain tissue segmentation" Part 13 of 14
<p>This dataset is part of the work <a href="https://www.frontiersin.org/articles/10.3389/fnimg.2022.1012639/full">https://www.frontiersin.org/articles/10.3389/fnimg.2022.1012639/full</a>. This is the thirteenth part of 14 parts of the full dataset (13/14). It contains 3 sets of simulated T1 weighted brain volumes in 3 simulated scanning sequences of spin-echo. The parameters of simulated scanning sequences are respectively repetition time (TR) = 600ms, 700ms, 800ms, and echo time (TE) = 35ms. Under <strong>each</strong> simulated scanning sequence, there are 500 brain volumes.</p> <p>The segmentation labels for each tissue are contained in the first part which you may find at <a href="https://zenodo.org/record/7294916">https://zenodo.org/record/7294916</a></p> <p>The simulation process of this dataset involves two processes. The first is to simulate one brain under different simulated scanning sequences. For this, we use BrainWeb <a href="https://brainweb.bic.mni.mcgill.ca/cgi/bw/submit_request">https://brainweb.bic.mni.mcgill.ca/cgi/bw/submit_request</a>. In the custom setting, we use spin-echo and apply image artifact the same as the default setting of this page. The second process is to transform each simulated brain from BrainWeb to different anatomical shapes. We use Human Connectome Project (HCP) 1200 subject data <a href="https://www.humanconnectome.org/study/hcp-young-adult">https://www.humanconnectome.org/study/hcp-young-adult</a> and randomly select 500 brains as anatomical references.</p> <p>Other details of this dataset can be found at <a href="https://www.frontiersin.org/articles/10.3389/fnimg.2022.1012639/full">https://www.frontiersin.org/articles/10.3389/fnimg.2022.1012639/full</a> where the details of the data construction are discussed.</p> <p>All parts of the whole dataset can be found at:</p> <p>Part 1: <a href="https://zenodo.org/record/7294916">https://zenodo.org/record/7294916</a></p> <p>Part 2: <a href="https://zenodo.org/record/7389550">https://zenodo.org/record/7389550</a></p> <p>Part 3: <a href="https://zenodo.org/record/7390382">https://zenodo.org/record/7390382</a></p> <p>Part 4: <a href="https://zenodo.org/record/7390741">https://zenodo.org/record/7390741</a></p> <p>Part 5: <a href="https://zenodo.org/record/7391205">https://zenodo.org/record/7391205</a></p> <p>Part 6: <a href="https://zenodo.org/record/7393060">https://zenodo.org/record/7393060</a></p> <p>Part 7: <a href="https://zenodo.org/record/7393174">https://zenodo.org/record/7393174</a></p> <p>Part 8: <a href="https://zenodo.org/record/7393347">https://zenodo.org/record/7393347</a></p> <p>Part 9: <a href="https://zenodo.org/record/7394250">https://zenodo.org/record/7394250</a></p> <p>Part 10: <a href="https://zenodo.org/record/7394667">https://zenodo.org/record/7394667</a></p> <p>Part 11: <a href="https://zenodo.org/record/7394939">https://zenodo.org/record/7394939</a></p> <p>Part 12: <a href="https://zenodo.org/record/7395031">https://zenodo.org/record/7395031</a></p> <p>Part 13: <a href="https://zenodo.org/record/7395620">https://zenodo.org/record/7395620</a></p> <p>Part 14: <a href="https://zenodo.org/record/7395622">https://zenodo.org/record/7395622</a></p> <p> </p>
Datasets of "Influence of contrast and texture based image modifications on the performance and attention shift of U-Net models for brain tissue segmentation" Part 7 of 14
<p>This dataset is part of the work <a href="https://www.frontiersin.org/articles/10.3389/fnimg.2022.1012639/full">https://www.frontiersin.org/articles/10.3389/fnimg.2022.1012639/full</a>. This is the seventh part of 14 parts of the full dataset (7/14). It contains 3 sets of simulated T1 weighted brain volumes in 3 simulated scanning sequences of spin-echo. The parameters of simulated scanning sequences are respectively repetition time (TR) = 300ms, 400ms, 500ms, and echo time (TE) = 40ms. Under <strong>each</strong> simulated scanning sequence, there are 500 brain volumes.</p> <p>The segmentation labels for each tissue are contained in the first part which you may find at <a href="https://zenodo.org/record/7294916">https://zenodo.org/record/7294916</a></p> <p>The simulation process of this dataset involves two processes. The first is to simulate one brain under different simulated scanning sequences. For this, we use BrainWeb <a href="https://brainweb.bic.mni.mcgill.ca/cgi/bw/submit_request">https://brainweb.bic.mni.mcgill.ca/cgi/bw/submit_request</a>. In the custom setting, we use spin-echo and apply image artifact the same as the default setting of this page. The second process is to transform each simulated brain from BrainWeb to different anatomical shapes. We use Human Connectome Project (HCP) 1200 subject data <a href="https://www.humanconnectome.org/study/hcp-young-adult">https://www.humanconnectome.org/study/hcp-young-adult</a> and randomly select 500 brains as anatomical references.</p> <p>Other details of this dataset can be found at <a href="https://www.frontiersin.org/articles/10.3389/fnimg.2022.1012639/full">https://www.frontiersin.org/articles/10.3389/fnimg.2022.1012639/full</a> where the details of the data construction are discussed.</p> <p>All parts of the whole dataset can be found at:</p> <p>Part 1: <a href="https://zenodo.org/record/7294916">https://zenodo.org/record/7294916</a></p> <p>Part 2: <a href="https://zenodo.org/record/7389550">https://zenodo.org/record/7389550</a></p> <p>Part 3: <a href="https://zenodo.org/record/7390382">https://zenodo.org/record/7390382</a></p> <p>Part 4: <a href="https://zenodo.org/record/7390741">https://zenodo.org/record/7390741</a></p> <p>Part 5: <a href="https://zenodo.org/record/7391205">https://zenodo.org/record/7391205</a></p> <p>Part 6: <a href="https://zenodo.org/record/7393060">https://zenodo.org/record/7393060</a></p> <p>Part 7: <a href="https://zenodo.org/record/7393174">https://zenodo.org/record/7393174</a></p> <p>Part 8: <a href="https://zenodo.org/record/7393347">https://zenodo.org/record/7393347</a></p> <p>Part 9: <a href="https://zenodo.org/record/7394250">https://zenodo.org/record/7394250</a></p> <p>Part 10: <a href="https://zenodo.org/record/7394667">https://zenodo.org/record/7394667</a></p> <p>Part 11: <a href="https://zenodo.org/record/7394939">https://zenodo.org/record/7394939</a></p> <p>Part 12: <a href="https://zenodo.org/record/7395031">https://zenodo.org/record/7395031</a></p> <p>Part 13: <a href="https://zenodo.org/record/7395620">https://zenodo.org/record/7395620</a></p> <p>Part 14: <a href="https://zenodo.org/record/7395622">https://zenodo.org/record/7395622</a></p>
An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery: Dataset
<p>This archive contains code and data to go with the paper <em>*An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery*</em>.</p> <p> </p> <p>This archive contains geospatial data, as well as the code used to generate the geospatial data.</p> <p>The geospatial data consists of georeferenced polygons identifying areas which are covered by green roofs in London (GBR) generated from 2019 aerial imagery.</p> <p>The data is described in detail in the manuscript <em>*An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery*</em>. See abstract below.</p> <p> </p> <p>GeoJSON format:</p> <p>GeoJSON is a format for encoding geospatial data, see https://geojson.org/.</p> <p>GeoJSON can be read using GIS programs including ArcGIS, QGIS, OGR.</p> <p> </p> <p>Contents:</p> <p>`geospatial_data/buffered_polygons_2021.zip` a zip archive containing a geojson file. It is the estimated locations of green roofs in London in 2021 and is the main result, which can be opened in any GIS program after being unzipped.</p> <p>`geospatial_data/buffered_polygons_2019.zip` a zip archive containing a geojson file. It is the estimated locations of green roofs in London in 2019 and is a secondary result, which can be opened in any GIS program after being unzipped. The predictions were made with the same model as the 2021 results.</p> <p>`geospatial_data/labelled_area.zip` a zip archive containing a geojson file. Identifies the area which was hand-labelled.</p> <p>`geospatial_data/manual_2021.zip` a zip archive containing a geojson file. Manually labelled green roof from 2021 imagery.</p> <p>`geospatial_data/manual_2019.zip` a zip archive containing a geojson file. Manually labelled green roof from 2019 imagery.</p> <p>`segmentation_code` contains the code used to produce the segmentation from the aerial imagery.</p> <p>`analysis_code` contains the code used to produce the plots and tables for the paper.</p> <p> </p> <p>Imagery availability:</p> <p>Unfortunately the aerial imagery and building footprint data cannot be shared directly, as you will require the proper license. Both can be found at [Digimap](https://digimap.edina.ac.uk) provided your institution has the license.</p> <p> </p> <p>Abstract:</p> <p>Green roofs can mitigate heat, increase biodiversity, and attenuate storm water, giving some of the benefits of natural vegetation in an urban context where ground space is scarce. To guide the design of more sustainable and climate resilient buildings and neighbourhoods, there is a need to assess the existing status of green roof coverage and explore the potential for future implementation. Therefore, accurate information on the prevalence and characteristics of existing green roofs is needed, but this information is currently lacking. Segmentation algorithms have been used widely to identify buildings and land cover in aerial imagery. Using a machine-learning algorithm based on U-Net to segment aerial imagery, we surveyed the area and coverage of green roofs in London, producing a geospatial dataset \cite[]{simpson_charles_2022_6861929}. We estimate that there was 0.23 km^2 of green roof in the Central Activities Zone (CAZ) of London, (1.07 km^2) in Inner London, and (1.89 km^2) in Greater London in the year 2021. This corresponds to 2.0% of the total building footprint area in the CAZ, and 1.3% in Inner London. There is a relatively higher concentration of green roofs in the City of London, covering 3.9% of the total building footprint area. Test set accuracy was 0.99, with an f-score of 0.58. When tested against imagery and labels from a different year (2019), the model performed just as well as a model trained on the imagery and labels from that year, showing that the model generalised well between different imagery. We improve on previous studies by including more negative examples in the training data, and by requiring coincidence between vector building footprints and green roof patches. We experimented with different data augmentation methods, and found a small improvement in performance when applying random elastic deformations, colour shifts, gamma adjustments, and rotations to the imagery. The survey covers 1558 km^2 of Greater London, making this the largest open automatic survey of green roofs in any city. The geospatial dataset is at the single-building level, providing a higher level of detail over the larger area compared to what was already available. This dataset will enable future work exploring the potential of green roofs in London and on urban climate modelling.</p>
Dataset for "Reproducibility and FAIR Principles: The Case of a Segment Polarity Network Model"
<p>Results of random sampling the segment polarity network with the simulator COPASI. These results correspond to Fig. 2 and Table 2 of von Dassow et. al (2000) (doi:10.1038/35018085). The random sampling was carried out with file vonDassow2000_1x4_alt.cps with COPASI version 4.39 selecting the appropriate parameter set named (1-7) and setting the number of repeats in the parameter scan task to the desired number. Full results of sampling are in files prefixed with the row number of Table 2 of von Dassow et. al (2000) and extension .tsv. Results with scores below 0.2 are in corresponding files with the word "-hits" in the filename. Includes also results from a time course simulation of this model using four different simulators (COPASI, Tellurium, Amici, and VCell). Finally also contains a study on multistability carried out by random sampling of parameters and initial conditions (run with COPASI). Markdown file README.md contains more detailed explanation. See also https://github.com/pmendes/models/tree/main/vonDassow2000</p>
replicAnt - Plum2023 - Segmentation Datasets and Trained Models
<p>This dataset contains all recorded and hand-annotated as well as all synthetically generated data as well as representative trained networks used for semantic and instance segmentation experiments in the<em> replicAnt - generating annotated images of animals in complex environments using Unreal Engine</em> manuscript. Unless stated otherwise, all 3D animal models used in the synthetically generated data have been generated with the open-source photgrammetry platform scAnt <a href="http://peerj.com/articles/11155/">peerj.com/articles/11155/</a>. All synthetic data has been generated with the associated replicAnt project available from <a href="https://github.com/evo-biomech/replicAnt">https://github.com/evo-biomech/replicAnt</a>.</p> <p><strong>Abstract:</strong></p> <p>Deep learning-based computer vision methods are transforming animal behavioural research. Transfer learning has enabled work in non-model species, but still requires hand-annotation of example footage, and is only performant in well-defined conditions. To overcome these limitations, we created replicAnt, a configurable pipeline implemented in Unreal Engine 5 and Python, designed to generate large and variable training datasets on consumer-grade hardware instead. replicAnt places 3D animal models into complex, procedurally generated environments, from which automatically annotated images can be exported. We demonstrate that synthetic data generated with replicAnt can significantly reduce the hand-annotation required to achieve benchmark performance in common applications such as animal detection, tracking, pose-estimation, and semantic segmentation; and that it increases the subject-specificity and domain-invariance of the trained networks, so conferring robustness. In some applications, replicAnt may even remove the need for hand-annotation altogether. It thus represents a significant step towards porting deep learning-based computer vision tools to the field.</p> <p><strong>Benchmark data</strong></p> <p>Semantic and instance segmentation is used only rarely in non-human animals, partially due to the laborious process of curating sufficiently large annotated datasets. <em>replicAnt </em>can produce pixel-perfect segmentation maps with minimal manual effort. In order to assess the quality of the segmentations inferred by networks trained with these maps, semi-quantitative verification was conducted using a set of macro-photographs of <em>Leptoglossus zonatus</em> (Dallas, 1852) and <em>Leptoglossus phyllopus</em> (Linnaeus, 1767), provided by Prof. Christine Miller (University of Florida), and Royal Tyler (Bugwood.org. For further qualitative assessment of instance segmentation, we used laboratory footage, and field photographs of <em>Atta vollenweideri</em> provided by Prof. Flavio Roces. More extensive quantitative validation was infeasible, due to the considerable effort involved in hand-annotating larger datasets on a per-pixel basis.</p> <p><strong>Synthetic data</strong></p> <p>We generated two synthetic datasets from a single 3D scanned <em>Leptoglossus zonatus</em> (Dallas, 1852) specimen: one using the default pipeline, and one with additional plant assets, spawned by three dedicated scatterers. The plant assets were taken from the Quixel library and include 20 grass and 11 fern and shrub assets. Two dedicated grass scatterers were configured to spawn between 10,000 and 100,000 instances; the fern and shrub scatterer spawned between 500 to 10,000 instances. A total of 10,000 samples were generated for each sub dataset, leading to a combined dataset comprising 20,000 image render and ID passes. The addition of plant assets was necessary, as many of the macro-photographs also contained truncated plant stems or similar fragments, which networks trained on the default data struggled to distinguish from insect body segments. The ability to simply supplement the asset library underlines one of the main strengths of <em>replicAnt</em>: training data can be tailored to specific use cases with minimal effort.</p> <p><strong>Funding</strong></p> <p>This study received funding from Imperial College’s President’s PhD Scholarship (to Fabian Plum), and is part of a project that has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant agreement No. 851705, to David Labonte). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.</p>
ISLES 2022: A multi-center magnetic resonance imaging stroke lesion segmentation dataset.
<p>This multi-center dataset consists of 250 expert-annotated magnetic resonance imaging stroke cases. It is the training dataset for the Ischemic Stroke Lesion Segmentation Challenge (ISLES'22).</p> <p>For each case, an expert level annotation of the stroke lesions is included along with the following three imaging sequences: Fluid attenuated inversion recovery (FLAIR), diffusion weighted imaging (DWI, b=1000) and its corresponding apparent diffusion coefficient (ADC) map. All imaging data and annotations are released in the Neuroimaging Informatics Technology Initiative (NIfTI) format (https://nifti.nimh.nih.gov/nifti-1), according to the BIDS convention. All imaging data are released in the native space without prior registration. Prior to release, skull-stripping was performed to de-identify patients.</p> <p>Image acquisition was performed on one of the following devices: 3T Philips MRI scanners (Achieva, Ingenia), 3T Siemens MRI scanner (Verio) or 1.5T Siemens MAGNETOM MRI scanners (Avanto, Aera). All images were obtained by healthcare professionals as part of the clinical imaging routine for stroke patients at three different stroke centers and imaging data was collected retrospectively for different clinical studies. Computer-readable scanner metadata from the Digital Imaging and Communications in Medicine (DICOM) header in the JSON file format is provided with the datasets if available.</p> <p>For a full dataset description, see the <a href="https://arxiv.org/abs/2206.06694">ISLES'22 preprint</a>.</p> <p>More information about the ISLES'22 challenge can be found in <a href="https://isles22.grand-challenge.org/">grand challenge</a> and in our official <a href="http://www.isles-challenge.org/">challenge website</a>.</p> <h3>Please cite the following works when using this dataset:</h3> <ul> <li>de la Rosa, Ezequiel, et al. <strong>DeepISLES: a clinically validated ischemic stroke segmentation model from the ISLES'22 challenge.</strong> <em>Nature Communications</em> 16.1 (2025): 7357.</li> <li>Hernandez Petzsche, Moritz R., et al. <strong>ISLES 2022: A multi-center magnetic resonance imaging stroke lesion segmentation dataset.</strong> <em>Scientific data</em> 9.1 (2022): 762.</li> </ul>
NPM3D dataset with instance labels used in paper "Toward Accurate Instance Segmentation in Large-scale LiDAR Point Clouds"
<p>NPM3D (https://npm3d.fr/paris-carla-3d) consists of mobile laser scanning (MLS) point clouds collected in four different regions in the French cities of Paris and Lille, where each point has been annotated with two labels: one that assigns it to one out of 10 semantic categories and another one that assigns it to an object instance. When inspecting the data, we found 9 cases where multiple tree instances had not been separated correctly (i.e., they had the same ground truth instance label). These cases were manually corrected using the CloudCompare software (https://www.cloudcompare.org), and 35 individual tree instances were obtained. Our variant of the dataset with 10 semantic categories and enhanced instance labels is publicly available.</p>
Parhyale 3D segmentation dataset
<p>The Parhyale 3D Segmentation dataset consists of 50 timepoints (TP01-TP50) of 3D images (512x512x34), where the manual annotations can be found at discrete 6 timepoints (at TP01, TP11, TP21, TP31, TP41 and TP50).</p> <p>For further details, see README file.</p> <p>This version fixes the duplicated label IDs found in the previous version of label files. This version ensures that each instance has a unique ID. Thanks to Jackson Borchardt for reporting that error.</p>
PTX-498: A multi-center pneumothorax segmentation chest X-ray image dataset
<p>Pneumothorax is a common medical emergency defined as the abnormal collection of air in the pleural space between the lung and chest wall. Its typical symptoms include chest pain and dyspnea, leading to oxygen deficiency or even life-threatening in severe cases. Therefore, an efficient and automatic pneumothorax diagnosis algorithm would be useful in many clinical scenarios. Recently, deep learning methods have achieved impressive progress in medical image segmentation tasks. However, a large-scale dataset is one of the critical components for the success of deep learning. On the other hand, there are few public chest X-ray images with pneumothorax.</p> <p>To stimulate the researchers' interest in the pneumothorax diagnosis algorithm, <strong>we released a new data set PTX-498 here. It contains 498 chest X-ray images of pneumothorax collected from three hospitals, and each image contains pixel-level annotations.</strong> All images were resized to 1024×1024. The raw image intensity was clipped according to the window width and level inside the dicom tag and then normalized to 0 to 255. The contours of the pneumothorax area were labelled by two senior radiologists using ITK-SNAP. The dataset was anonymized and every record related to patients' privacy was removed. Only the image data and the corresponding labels were included in PTX-498.</p> <p><strong>Please use the latest v2-fix version which removes duplicate images and uses the window width and level from the original dicom tag for normalization.</strong></p> <p><strong>Citation: If you are interested in this dataset and applying it in your research, please cite the following article.</strong><br> Paper link: https://doi.org/10.1016/j.neucom.2021.05.029<br> Cite this article as Yunpeng Wang, Kang Wang, Xueqing Peng, Lili Shi, Jing Sun, Shibao Zheng, Fei Shan, Weiya Shi, Lei Liu*. DeepSDM: Boundary-aware pneumothorax segmentation in chest X-ray images [J]. Neurocomputing, 2021, 454: 201-211.</p> <div> <div class="gtx-trans-icon"> </div> </div>
Dataset for Detection and Segmentation of the Radiographic Features of Pulmonary Edema
<p><strong>Objectives:</strong> This comprehensive dataset is well suited for training, evaluating, and using machine learning models to detect, segment, and analyze radiological features associated with pulmonary edema in chest X-ray images.</p> <p><strong>Description:</strong> This dataset consists of a collection of chest X-rays extracted from the <a href="https://physionet.org/content/mimic-cxr-jpg/2.0.0/" target="_blank" rel="noopener">MIMIC database</a>, carefully collected at the Beth Israel Deaconess Medical Center. In total, it comprises 1000 chest X-rays obtained from 741 patients with features suggestive of edema. These X-rays were carefully selected for manual annotation. The annotations are rich and detailed, covering specific radiological features commonly associated with pulmonary edema, including cephalization, Kerley lines, pleural effusions, bat wings, and infiltrates. The dataset includes a wide variety of radiological features, with a total of 4263 annotations (<em>Table 1</em>). Furthermore, each chest radiograph is thoughtfully assigned a severity category, categorizing it as "no edema", "vascular congestion", "interstitial edema", or "alveolar edema".</p> <p><strong>Annotation Method:</strong> The annotation process was meticulously performed by a highly qualified clinician with over 10 years of radiology experience, utilizing both frontal and lateral views for each chest X-ray study. Cephalization and Kerley lines were delineated using polylines, while other features were delineated using binary masks. This methodological approach was carefully chosen to provide a comprehensive data set that would ensure accuracy in subsequent analyses and label assignments. </p> <p>Notably, all features are represented as bounding boxes, meticulously defined by their respective upper-left (x1; y1) and lower-right (x2; y2) corners. In addition, selected features are provided with masks encoded in base 64 format. To facilitate seamless decoding, we provide a conversion script called "mask_converter.py" that allows the transformation of encoded masks into a versatile numpy array format. This feature improves the usability of the dataset for precise analysis and deep learning applications.</p> <p><strong>Datasets:</strong></p> <ol> <li><strong>SLY dataset:</strong> The dataset contains chest X-ray images labeled by clinicians, including both stacked frontal and lateral images. We obtained this dataset by annotating it on the <a href="https://supervisely.com/" target="_blank" rel="noopener">Supervisely platform</a>, and it is stored in JSON and PNG formats.</li> <li><strong>Source dataset:</strong> The dataset is a transformed version of the SLY dataset. In this dataset, all annotations are consolidated into a single spreadsheet, and only frontal view images are represented.</li> <li><strong>Processed dataset: </strong>The dataset focuses exclusively on the lung area for analysis, as other areas surrounding the lung typically contain extraneous information that clinicians do not use in their decision-making process.</li> <li><strong>COCO dataset:</strong> A collection of subsets prepared in the <a href="https://towardsdatascience.com/how-to-work-with-object-detection-datasets-in-coco-format-9bf4fb5848a4" target="_blank" rel="noopener">COCO format</a> and suitable for training and testing. It includes subsets for each feature and for all features evaluated in this study.</li> </ol> <div> <p><strong>Access to the Study:</strong> Further information about this study, including curated source code, dataset details, and trained models, can be accessed through the following repositories:</p> <ul> <li><strong>Source code:</strong> <a href="https://github.com/ViacheslavDanilov/edema_quantification" target="_blank" rel="noopener">https://github.com/ViacheslavDanilov/edema_quantification</a></li> <li><strong>Dataset:</strong> <a href="https://doi.org/10.5281/zenodo.8383776" target="_blank" rel="noopener">https://doi.org/10.5281/zenodo.8383776</a></li> <li><strong>Lung segmentation models:</strong> <a href="https://doi.org/10.5281/zenodo.8393555" target="_blank" rel="noopener">https://doi.org/10.5281/zenodo.8393555</a></li> <li><strong>Radiographic feature detection models:</strong> <a href="https://doi.org/10.5281/zenodo.8393565" target="_blank" rel="noopener">https://doi.org/10.5281/zenodo.8393565</a></li> </ul> <p> </p> <p><em><strong>Table 1.</strong> Summary of annotated radiological features and severity labels</em></p> <table> <tbody> <tr> <td> <p><strong>Radiological feature</strong></p> </td> <td> <p><strong>Number </strong><strong>of objects</strong></p> </td> <td> <p><strong>Severity </strong><strong>l</strong><strong>abel</strong></p> </td> <td> <p><strong>Number of cases</strong></p> </td> </tr> <tr> <td> <p>Cephalization</p> </td> <td> <p>1656</p> </td> <td> <p>No edema</p> </td> <td> <p>21</p> </td> </tr> <tr> <td> <p>Kerley line</p> </td> <td> <p>609</p> </td> <td> <p>Vascular congestion</p> </td> <td> <p>74</p> </td> </tr> <tr> <td> <p>Pleural effusion</p> </td> <td> <p>317</p> </td> <td> <p>Interstitial edema</p> </td> <td> <p>51</p> </td> </tr> <tr> <td> <p>Bat wing</p> </td> <td> <p>1604</p> </td> <td> <p>Alveolar edema</p> </td> <td> <p>595</p> </td> </tr> <tr> <td> <p>Infiltrate</p> </td> <td> <p>77</p> </td> <td> <p> </p> </td> <td> <p> </p> </td> </tr> <tr> <td> <p>TOTAL</p> </td> <td> <p>4263</p> </td> <td> <p>TOTAL</p> </td> <td> <p>741</p> </td> </tr> </tbody> </table> <p> </p> </div>
A multi-scale labeled dataset for boulder segmentation and navigation on small bodies
<p>The capability to detect boulders on the surface of small bodies is beneficial for vision-based applications such as hazard detection during critical operations, safety quantification, autonomous planning of scientific operations, and autonomous navigation. This task, however, is challenging due to the wide assortment of irregular shapes, the characteristics of the boulders population, and the rapid variability in the illumination conditions. Moreover, the lack of publicly available labeled datasets damps the research about data-driven algorithms. The following dataset has been designed and made publicly available to tackle these challenges. Its purpose is twofold. First, from the lessons learned from previous datasets, to develop a multi-purpose, high-fidelity dataset with boulders scattered across the surface of a small body. Second, to exploit domain randomization, artificial noise addition, scaling, and post-processing, enabling the design of data-driven pipelines. </p> <p>The methodology used to generate the dataset is illustrated in the work "A multi-scale labeled dataset for boulder segmentation and navigation on small bodies" by Mattia Pugliatti and Michele Maestrini, presented at the 74th IAC (International Astronautical Congress), 2024, Baku, Azerbaijan.</p> <p>The dataset contains the image-label pairs of 47502 samples, organized with the following structure: </p> <p>Dataset_PugliattiMaestrini_2023IAC<br> --img<br> --labels<br> --masks</p> <p>The dataset is comprised of 47502 samples. The "img" folder contains the input, 512x 512 grayscale images. The "labels" folder includes the .txt segmentation labels of the 15 most prominent boulders for each image detected with the methodology illustrated in the IAC paper. The "masks" dataset contains the segmentation masks for all image layers, with the values being encoded between 0 and 17 as uint8. The samples are named as XXXXXX_YYY. XXXXXX stands for the image's original ID during rendering. YYY corresponds to the sub-splits of the original image obtained at rendering: </p> <p> 001 - Top-Left crop<br> 002 - Top-Right crop<br> 003 - Bottom-Left crop<br> 004 - Bottom-right crop<br> 005 - Whole, resized</p> <p>The file "10000_ub_2023-01-18 00.09.43.txt" contains all the values of the rendering inputs detailed in the IAC paper.</p>
Counted Biopores dataset used in 'RootPainter: Deep Learning Segmentation of Biological Images with Corrective Annotation'
<p>Counted biopores dataset used in the article: 'RootPainter: Deep Learning Segmentation of Biological Images with Corrective Annotation'</p> <p>Originally collected as part of a field trial at the University of Bonn in 2012, described in the following paper:</p> <p>Eusun Han, Timo Kautz, Ute Perkons, Marcel Lüsebrink, Ralf Pude, and Ulrich Köpke.Quantification of soil biopore density after perennial fodder cropping.Plant and Soil, 394(1-2):73–85, sep 2015. ISSN 15735036. doi:10.1007/s11104- 015- 2488- 3</p>
Counted Nodules dataset used in 'RootPainter: Deep Learning Segmentation of Biological Images with Corrective Annotation'
<p>Counted Nodules dataset used in the article: 'RootPainter: Deep Learning Segmentation of Biological Images with Corrective Annotation'</p>
Training dataset for semantic segmentation (U-Net) of structural conservation practices
<p>In this research, the best management practices include vegetative/structural conservation practices (SCP) across crop fields, such as grassed waterways and terraces. This reference dataset includes 500,000 pair patches (false-color image (B1: NIR, B2: Red, B3: Green) and binary label (SCP: yes[1] or no[0]). These training samples were randomly extracted from Iowa BMP project (<a href="https://www.gis.iastate.edu/gisf/projects/conservation-practices">https://www.gis.iastate.edu/gisf/projects/conservation-practices</a>) and present 90% of patches with SCP areas and 10% of patches non-SCP area. The patch dimension is 256 x 256 pixels at 2-m resolution. Due to the file size, the images were upload in different *.rar files (imagem_0_200k.rar, imagem_200_400k.rar, imagem_400_500k.rar), and the user should download all and merge them in the same folder. The corresponding labels are all in "class_bin.rar" file.</p> <p>Application: These pair images are useful for conservation practitioners interested in the classification of vegetative/structural SCPs using deep-learning semantic segmentation methods.</p> <p>Further information will be available in future.</p>
Dataset for semantic segmentation of the laboratory model of manufacturing environment
<p>This dataset includes images and labels used for semantic segmentation of the laboratory model of the manufacturing environment, at the University of Belgrade - Faculty of Mechanical Engineering. The dataset is gathered by using mobile robot RAICO (Robot with Artificial Intelligence based COgnition) and its stereo visual system made from two Basler acA1920-25uc cameras with Fujinon lens DF6HA-1B. The dataset includes close to 430 images with a resolution of 640x360. Images are acquired by both cameras at different mobile robot poses in the laboratory model of a manufacturing environment. Five classes are introduced in the dataset, machines 1 to 4, and a background class. The exact names of the classes are:</p> <p>classNames = ["Machine_1", "Machine_2", "Machine_3", "Machine_4", "Background"];</p> <p>while the labels of the classes (RGB values of label images) are:</p> <p>labelIDs = [ ...<br> 000 000 255; ... % "Machine 1"<br> 000 255 255; ... % "Machine 2"<br> 255 255 000; ... % "Machine 3"<br> 255 000 000; ... % "Machine 4"<br> 255 255 255; ... % "Background"<br> ];</p> <p>Image and label pairs are entitled 1 to 430, and e.g. label 5 corresponds to images 5.</p> <p>This dataset was developed with the support of the Science Fund of the Republic of Serbia, Grant No. 6523109, AI - MISSION4.0, 2020-2022.</p>
Dataset and Jupyter worksheet interpreting the (results from) small- and wide-angle scattering data from a series of boehmite/epoxy nanocomposites. Accompanies the publication "Competition of nanoparticle-induced mobilization and immobilization effects on segmental dynamics of an epoxy-based nanocomposite"
<p>Dataset and Jupyter worksheet interpreting the (results from) small- and wide-angle scattering data from a series of boehmite/epoxy nanocomposites. Accompanies the publication "Competition of nanoparticle-induced mobilization and immobilization effects on segmental dynamics of an epoxy-based nanocomposite", by Paulina Szymoniak, Brian R. Pauw, Xintong Qu, and Andreas Schönhals.</p> <p>Datasets are in three-column ascii (processed and azimuthally averaged data) from a Xenocs NanoInXider SW instrument. Monte-Carlo analyses were performed using McSAS 1.3.1, other analyses are in the Python 3.7 worksheet. Graphics and result tables are output by the worksheet. </p>
Turkish Makam Symbolic Phrase Segmentation Dataset
<p>makam-symbolic-phrase-segmentation-dataset</p> <p><strong>Data sets containing pieces in SymbTr2 format segmented into phrases</strong></p> <p>This study presents a large machine-readable dataset of Turkish makam music scores segmented into phrases by experts of this music. The segmentation facilitates computational research on melodic similarity between phrases, and relation between melodic phrasing and meter, rarely studied topics due to unavailability of data resources. It consists of 31362 phrases on a set of 480 scores of different compositions annotated by 3 experts.</p> <p>Please refer to the following publication if you use this data in your research:</p> <blockquote> <p>M. K. Karaosmanoglu, B. Bozkurt, A. Holzapfel, N. D. Disiacik, A symbolic dataset of Turkish makam music phrases, Folk Music Analysis Workshop (FMA), Istanbul, 2014.</p> </blockquote> <p>The refactored code for automatic phrase segmentation can be found here.</p> <p>For other deliverables of the paper please visit: http://www.rhythmos.org/shareddata/turkishphrases.html</p>
Jingju a cappella singing pitch contour segmentation ground truth dataset
<p>The dataset used in the paper:</p> <blockquote> <p>Gong, Rong; Yang, Yile; Serra, Xavier; Pitch Contour Segmentation for Computer-aided Jingju Singing Training Sound and Music Computing (SMC 2016), 2016, Hamburg, Germany</p> </blockquote> <p>is in "dataset" folder. The a cappella singing audio recordings are not contained in this folder due to their large size, please contact the paper authors to request them (rong.gong@upf.edu). In the "dataset" folder you can find:</p> <ol> <li>ground truth</li> <li>Jinging singing scores in .xml format used for estimating the bigram note transition probabilities.</li> </ol> <p>The ground truth annotation is used for:</p> <ul> <li>melodic transcription (male_12_pos_1 missing)</li> <li>parameter optimization,</li> <li>evaluating the StdCdLe thresholding and the overall segmentation performance.</li> </ul> <p>The subfolder "groundtruth" contains the following annotation for each jingju a cappella audio:</p> <ul> <li>file name: description (format)</li> <li>*_melodicTrans.csv: melodic transcription ground truth used for the evaluation (start_time pitch duration -).</li> <li>*_coarseSeg.csv: StdCdLe ground truth used for the parameter optimization and the evaluation (segmentation points).</li> <li>*_refinedSeg.csv: ground truth used for optimizing other parameters and the evaluation (start_time - duration).</li> <li>*_pitchtrack.csv: pitch track (contour) extracted by pYIN pitch-tracking algorithm (filename time pitch).</li> <li>*_monoNoteOut.csv: notes estimated by pYIN note-tracking algorithm (filename start_time duration pitch).</li> </ul> <p> </p> <p> </p>
CytoNuke Dataset: Towards reliable whole-cell segmentation in bright-field histological images
<p>This is the dataset from the preprint "Cyto R-CNN and CytoNuke Dataset: Towards reliable whole-cell segmentation in bright-field histological images" by Raufeisen et al. (2024). It contains 6,683 annotations (3,991 nuclei and 2,607 whole cells) of head and neck squamous cell carcinoma cells in hematoxylin and eosin stained histological images. The annotations are in COCO format and distributed over 83 PNG images. Cyto R-CNN was trained on this dataset and compared with other state-of-the-art methods. The CytoNuke dataset is released under the CC BY 4.0 license.</p> <p>The histological images are from the CPTAC dataset:<br>National Cancer Institute Clinical Proteomic Tumor Analysis Consortium (CPTAC). (2018). The Clinical Proteomic Tumor Analysis Consortium Head and Neck Squamous Cell Carcinoma Collection (CPTAC-HNSCC) (Version 15) [Data set]. The Cancer Imaging Archive. https://doi.org/10.7937/K9/TCIA.2018.UW45NH81</p> <p>Funding: Behrus Puladi was funded by the Medical Faculty of RWTH Aachen University as part of the Clinician Scientist Program. We acknowledge FWF enFaced 2.0 [KLI 1044, https://enfaced2.ikim.nrw/] and KITE (Plattform für KI-Translation Essen) from the REACT-EU initiative [https://kite.ikim.nrw/, EFRE-0801977]. Fabian Hörst, Jianning Li, Jens Kleesiek and Jan Egger received funding from the Cancer Research Center Cologne Essen (CCCE).</p>
Dataset with segmentations of 117 important anatomical structures in 1228 CT images
<p>Info: This is version 2 of the TotalSegmentator dataset.<br><br>In 1228 CT images we segmented 117 anatomical structures covering a majority of relevant classes for most use cases. The CT images were randomly sampled from clinical routine, thus representing a real world dataset which generalizes to clinical application. The dataset contains a wide range of different pathologies, scanners, sequences and institutions.</p><p>Link to a copy of this dataset on Dropbox for much quicker download: <a href="https://www.dropbox.com/scl/fi/oq0fsz8oauory204g8o6f/Totalsegmentator_dataset_v201.zip?rlkey=afnl2ixhqca2ukkf1v9p6jz7p&dl=0">Dropbox Link</a></p><p>Overview of differences to v1 of this dataset: <a href="https://github.com/wasserth/TotalSegmentator/blob/master/resources/improvements_in_v2.md">here</a></p><p>A small subset of this dataset with only 102 subjects for quick download+exploration can be found here: <a href="https://doi.org/10.5281/zenodo.8367169">here</a></p><p>You can find a segmentation model trained on this dataset <a href="https://github.com/wasserth/TotalSegmentator">here</a>.<br><br>More details about the dataset can be found in the corresponding <a href="https://doi.org/10.1148/ryai.230024">paper</a> (the paper describes v1 of the dataset). Please cite this paper if you use the dataset.</p><p>This dataset was created by the department of <a href="https://www.unispital-basel.ch/en/radiologie-nuklearmedizin/forschung-radiologie-nuklearmedizin">Research and Analysis at University Hospital Basel</a>.</p><p><strong>UPDATE</strong>: On 2023-10-27 we uploaded version 2.0.1 which fixes broken files.</p>
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.