Find research datasets worth reusing

opencc-by-3.0Apr 2016View details →

Fig. 7. Distal flagellar segments. A in The British species of Enicospilus (Hymenoptera: Ichneumonidae: Ophioninae)

Fig. 7. Distal flagellar segments. A. Enicospilus adustus (Haller, 1885). B. E. cerebrator Aubert, 1966.

opencc-by-4.0Oct 2020View details →

Dataset for semantic segmentation of the laboratory model of manufacturing environment

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: classNames = ["Machine_1", "Machine_2", "Machine_3", "Machine_4", "Background"]; while the labels of the classes (RGB values of label images) are: labelIDs = [ ...     000 000 255; ... % "Machine 1"     000 255 255; ... % "Machine 2"     255 255 000; ... % "Machine 3"     255 000 000; ... % "Machine 4"     255 255 255; ...   % "Background"     ]; Image and label pairs are entitled 1 to 430, and e.g. label 5 corresponds to images 5. This dataset was developed with the support of the Science Fund of the Republic of Serbia, Grant No. 6523109, AI - MISSION4.0, 2020-2022.

opencc-by-4.0Dec 2020View details →

Semantic Segmentation of Time Series Imagery Using Deep Convolutional Neural Networks: A Case Study of Sandbars in Grand Canyon

This dataset contains imagery used to train and test Deep Convolutional Neural Networks for the purpose of binary semantic segmentation of a time series of oblique imagery capturing sandbar monitoring sites in The Grand Canyon. In addition the scripts needed for removing image distortion, registering, rectifying, and labeling imagery is present.

opencc-by-4.0Dec 2020View details →

Ilastik_organoid_segmentation

This repository holds two Ilastik projects and respective training data used for the organoid segmentation of the initial organoid screen in the publication "A semi-automated intestinal organoid screening method demonstrates epigenetic control of epithelial differentiation" (<a href="https://doi.org/10.3389/fcell.2020.618552">Ostrop et al. 2020</a>). The image pre-processing was carried out with the ImageJ script available on GitHub: <a href="https://github.com/jennyostrop/Fiji_organoid_brightfield_processing">https://github.com/jennyostrop/Fiji_organoid_brightfield_processing</a>, archived on Zenodo as <a href="https://doi.org/10.5281/zenodo.3951125">https://doi.org/10.5281/zenodo.3951125</a> For training, 1 well treated with DMSO vehicle control of 4 biological replicates and 5 timepoints were used (total 20 images). Training data was excluded from further analysis. The raw images were deposited to the Image Data Resource (<a href="https://idr.openmicroscopy.org">https://idr.openmicroscopy.org</a>) under accession number idr0092. Training data corresponds to Well A01 in plate 1-plate 4.   Project 1: Ilastik1_PixelClass_Edges Pixel classification, created with Ilastik 1.3.2 Input Raw Data: Summary projection of Sobel edge detection for each z-layer (Output 2d from ImageJ script) Training classes: Background: lines over background and debris (size 7) and inside of organoids (size 3) Object: exact lines following the outer border of organoids (size 1)   Project 2: Ilastik2_ObjectClass_Projection Object classification [Inputs: Raw Data, Pixel Prediction Map], created with Ilastik 1.3.2 Input Raw Data: Minimum intensity projections of each Zstack (Output 1b from ImageJ script) Input Prediction Maps: Pixel Prediction Maps generated in project Ilastik1_PixelClass_Edges Threshold and Size Filter: Method - Simple, Input - 0 (Background), Smooth – 1.0, 1.0, Threshold  - 0.55, Size Filter – Min 10, Max 1000000 Object classes: Mislabelled: Patch of background enclosed by objects, light Organoid: Typical organoid, medium dark Sphere_big: Sphere, almost round, light, size of organoids or bigger Sphere_small: Sphere, almost round, light to medium dark, small Cluster: Cluster of several organoids/spheres and organoids segmented as single object Debris: Debris in Matrigel, irregular borders and structure, mostly small, medium dark AirBubble: Air bubbles in Matrigel at early timepoints, round, dark with light centre Edges: Edges of well plate (empty class, included for consistency)

opencc-by-4.0Dec 2020View details →

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"

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. 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.

opencc-by-4.0Jan 2021View details →

Experiment data in support of "Segmentation analysis and the recovery of queuing parameters via the Wasserstein distance: a study of administrative data for patients with chronic obstructive pulmonary disease"

This archive contains a ZIP archive, `data.zip`, that itself contains the data used in the final sections of the paper. The remainder of the paper's supporting files are available at <a href="https://github.com/daffidwilde/copd-paper/">github.com/daffidwilde/copd-paper/</a> The ZIP archive is structured as follows: <ul> <li>There is a directory, `wasserstein`, for the parameter sweep described in the model construction section of the paper. Its contents are: (i) a file, `main.csv`, describing each parameter and their maximal Wasserstein distance to the observed data, and (ii) three directories, `best`, `median` and `worst`, each containing the simulated queuing results (in `main.csv`) from that sweep with the best, median and worst found parameter sets, respectively (in `params.txt`).</li> <li>The remaining three directories correspond to the experiments conducted in the final section of the paper. Each directory contains two files: (i) `system_times.csv` which holds trial parameters and system time records for every patient to pass through the model in that experiment, and (ii) `utilisations.csv` which holds trial parameters and utilisations for each server in the model for that experiment.</li> </ul>

opencc-by-4.0Jan 2021View details →

Human and Mouse Eyes for Pupil Semantic Segmentation

A dataset composed of 11897 grayscale images of humans (4285) and mouse (7612) eyes. In different experimental conditions:  head-fixation sessions (HF: 5061), 2-photon Ca2+ imaging ( 2P: 2551), and human eyes (H: 4285). The dataset contains 1596 eye blinks, 841 images in the mouse, and 755 photos in the human datasets. Five human raters segmented the pupil in all pictures (one per image) by manual placement of an ellipse or polygon over the pupil area. Raters flagged blinks using the same code.  All the photos are illuminated using infrared (IR, 850 nm) light sources. The dataset contains 2 folders: 'fullFrames': contains all the grayscale images in png format. 'annotation': contains a folder called 'png' with pupil mask in the red channel. There is also a file called 'annotations.csv' containing a list with a description of each file in the dataset in this folder. Description of the fields in annotations.csv: filename: [string] with the file name  eye: [0,1] if true an eye is present in the picture blink: [0,1] if true the subject is blinking exp: [string] what kind of experiments  w: [int] resolution width h: [int] resolution height roi_x: [int] roi x coordinate roi_y: [int] roi y coordinate roi_w: [int] roi width-height (128x128) sub: [int] subject's label

Figs 22–23. Thorax and abdominal segments 4–5 in Comparative morphology of immature Trictenotoma formosana Kriesche, 1919 and systematic position of the Trictenotomidae (Coleoptera, Tenebrionoidea)

Figs 22–23. Thorax and abdominal segments 4–5 of first instar larva of Trictenotoma formosana Kriesche, 1919. 22. Dorsal view. 23. Ventral view. Scale bars: 0.2 mm.

Figs 24–25. Abdominal segments 8–9 in Comparative morphology of immature Trictenotoma formosana Kriesche, 1919 and systematic position of the Trictenotomidae (Coleoptera, Tenebrionoidea)

Figs 24–25. Abdominal segments 8–9 of first instar larva of Trictenotoma formosana Kriesche, 1919. 24. Dorsal view. 25. Ventral view. Scale bars: 0.1 mm.

Figs 11–17. Thorax and abdominal segments 5 and 9 in Comparative morphology of immature Trictenotoma formosana Kriesche, 1919 and systematic position of the Trictenotomidae (Coleoptera, Tenebrionoidea)

Figs 11–17. Thorax and abdominal segments 5 and 9 of last-instar larva of Trictenotoma formosana Kriesche, 1919. 11–12. Thorax (11 = dorsal view, 12 = ventral view). 13. Tergite 5. 14. Sternite 5. 15. Sternite 9. 16. Tergite 9. 17. Abdominal segment 9, posterior view. Scale bars: 3 mm.

Figs 31–34. Abdominal segment 9 in Comparative morphology of immature Trictenotoma formosana Kriesche, 1919 and systematic position of the Trictenotomidae (Coleoptera, Tenebrionoidea)

Figs 31–34. Abdominal segment 9, spherical protrusions and spiracles of pupa of Trictenotoma formosana Kriesche, 1919. 31–33. Abdominal segment 9 (31 = dorsal view, 32 = lateral view, 33 = ventral view). 34. Spherical protrusions and spiracles on segments 4–6, lateral view. Abbreviations: sp = spiracle; lsp = lateral spherical protrusion; ssp = second spherical protrusion. Scale bars: 31–33 = 3 mm; 34 = 2 mm.

opencc-by-nc-sa-4.0Nov 2016View details →

Turkish Makam Symbolic Phrase Segmentation Dataset

makam-symbolic-phrase-segmentation-dataset Data sets containing pieces in SymbTr2 format segmented into phrases 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. Please refer to the following publication if you use this data in your research: <blockquote> 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. </blockquote> The refactored code for automatic phrase segmentation can be found here. For other deliverables of the paper please visit: http://www.rhythmos.org/shareddata/turkishphrases.html

opencc-by-4.0May 2016View details →

Validation data set for automatic blood vessel segmentation in colorectal cancer histology (IHC)

Content This data set contains 100 histological image patches of 1000 * 1000 px size. The samples were immunostained for CD34 (3,3'-Diaminobenzidine, DAB [brown]) with hematoxylin (blue) counterstain. Furthermore, the data set contains a table of blood vessel counts in each image by three blinded observers as well as an automatic count with a method based on the following paper: Kather, Jakob Nikolas et al. "Continuous Representation Of Tumor Microvessel Density And Detection Of Angiogenic Hotspots In Histological Whole-Slide Images". Oncotarget 6.22 (2015): 19163-19176. http://dx.doi.org/10.18632/oncotarget.4383 Image format All images are RGB, 0.50 µm per pixel, digitized with an Aperio ScanScope (Aperio/Leica biosystems), magnification 20x. Histological samples are fully anonymized images of formalin-fixed paraffin-embedded human colorectal adenocarcinomas (primary tumors and liver metastases) from our pathology archive (Institute of Pathology, University Medical Center Mannheim, Heidelberg University, Mannheim, Germany). Ethics statement All experiments were approved by the institutional ethics board (medical ethics board II, University Medical Center Mannheim, Heidelberg University, Germany; approval 2015-868R-MA). The institutional ethics board waived the need for informed consent for this retrospective analysis of anonymized samples. All experiments were carried out in accordance with the Declaration of Helsinki. Contact For questions, please contact: Dr. Jakob Nikolas Kather http://orcid.org/0000-0002-3730-5348 ResearcherID: D-4279-2015

opencc-by-nc-4.0Jul 2017View details →

Jingju a cappella singing pitch contour segmentation ground truth dataset

The dataset used in the paper: <blockquote> Gong, Rong; Yang, Yile; Serra, Xavier;  Pitch Contour Segmentation for Computer-aided Jingju Singing Training Sound and Music Computing (SMC 2016), 2016, Hamburg, Germany </blockquote> 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: <ol> <li>ground truth</li> <li>Jinging singing scores in .xml format used for estimating the bigram note transition probabilities.</li> </ol> The ground truth annotation is used for: <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> The subfolder "groundtruth" contains the following annotation for each jingju a cappella audio: <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>

opencc-by-4.0Jan 2024View details →

CytoNuke Dataset: Towards reliable whole-cell segmentation in bright-field histological images

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. The histological images are from the CPTAC dataset: 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 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).

Dataset with segmentations of 117 important anatomical structures in 1228 CT images

Info: This is version 2 of the TotalSegmentator dataset. 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.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>Overview of differences to v1 of this dataset: <a href="https://github.com/wasserth/TotalSegmentator/blob/master/resources/improvements_in_v2.md">here</a>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>You can find a segmentation model trained on this dataset <a href="https://github.com/wasserth/TotalSegmentator">here</a>. 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.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>.UPDATE: On 2023-10-27 we uploaded version 2.0.1 which fixes broken files.

opencc-by-4.0Oct 2023View details →

Vine Trunk Semantic Segmentation using Individual Vine Trunks

Automatically labeled semantic segmentation dataset of vine trunk thumbnails extracted from VIneSet and other supplementary data. Source paper of this dataset is "Generating vine trunk semantic segmentation dataset via semi-supervised learning and object detection" published in MDPI Robotics https://www.mdpi.com/2218-6581/13/2/20.

FIGURE 5 U in Automated segmentation of insect anatomy from micro-CT images using deep learning

FIGURE 5 U-Net implementation. The architecture of the used convolutional neural network (CNN) is an implementation of U-Net. It consists of two parts: two 3×3 convolutions followed by 2×2 max pooling and two 3×3 convolutions followed by 2×2 upconvolutions. Dropout was added to avoid overfitting. As a final step a 1×1 convolution is applied, resulting in an output map with two classes.

FIGURE 6 in Automated segmentation of insect anatomy from micro-CT images using deep learning

FIGURE 6 Network performance evaluation. High true positive rate (TPR) and low false positive rate (FPR) values for training (blue) and testing data (red) indicate the network's high generalizability.

FIGURE 10 in Automated segmentation of insect anatomy from micro-CT images using deep learning

FIGURE 10 Application of pipeline for other insect species. The brain textures of various insect species can be very similar to those of ants, facilitating the prediction by the network even without pretraining on specific insect brain scans. (a) Raw image of wasp head (original 1000 × 1000 px) and (b) its prediction without postprocessing (original 520 × 520 px), indicating satisfactory identification of the borders of the brain area. (c) 2D image of praying mantis head (520 × 520 px) and (d) the prediction of its brain area without postprocessing. Even though the network overpredicts some small pixel islands, it excludes from its prediction areas of the muscles, fibers, and cuticle.