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.
979
datasets available to search
ShareScore release 0.9.0
Dataset results
979 results for “Image dataset”
Dataset for "Learning to Read and Follow Music in Complete Score Sheet Images"
<p>This is the data used in the paper:<br> Florian Henkel , Rainer Kelz, Gerhard Widmer. "Learning to Read and Follow Music in Complete Score Sheet Images". Proceedings of the 21st International Society for Music Information Retrieval Conference (ISMIR), 2020</p> <p>The dataset is based on the <a href="https://zenodo.org/record/2597505#.XxbJtxGxVhE">MSMD - Multimodal Sheet Music Dataset</a> and processed for the purpose of score following in complete score sheet images.</p> <pre>Please find further information and the corresponding code on this Github page: <a href="https://github.com/CPJKU/audio_conditioned_unet">https://github.com/CPJKU/audio_conditioned_unet</a> </pre>
Multi-sensor dataset for testing merge of Hyperspectral, HD and 3D cloud information for image recognition
<p>This data contain multisensor image dataset constructed for benchmarking purposes. It contains multiple images of the constructed scenes -- on which objects made of different materials are placed to test image recognition scenarios. The scene is recorded from various angles by imagining sensors, i.e. HSI camera, HD camera on mobile chassis and MS Kinect to provide complete information.<br> </p> <p><strong>Equipment</strong><br> The imaging was performed with use of three devices for three different approaches to data. Those three devices' imaging characteristics are widely different when it comes to angle and resolution which required them to be separately positioned to acquire the matching images. Therefore while HD camera was being transferred on the moving platform (chassis), both Kinect and SOC710 were placed on a stationary position which was moved between the frames by hand.</p> <p><em>Hyperspectral data</em></p> <p><br> Hyperspectral data acquisition was performed with Surface Optics SOC710 camera. This camera records spectra at VNIR range $377-1046$ nm; the output image has dimensions $696 \times 520$ with 128 bands and $12$ bit dynamic range.</p> <p>The camera is equipped with sensor line translation unit and can be used from static stand as a conventional camera (i.e. it does not require mechanical translation of the observed sample or rotary stand, as in traditional ‘push broom’ hyperspectral cameras). The lighting was provided with four ambient lamps and adjusted for each scenario separately, so that most of the dynamic range of the camera was used and image saturation is avoided. Captured hyperspectral images were subject to a standard calibration procedure, including: the removal of a dark frame, spectral and radiometric calibration as well as reflectance normalization using the calibration panel. </p> <p><em>3D point clouds</em></p> <p><br> The Kinect sensor incorporates several advanced sensing hardware. The depth sensor consists of the IR projector combined with the IR camera, which is a monochrome complementary metaloxide semiconductor (CMOS) sensor. The IR projector is an IR laser that passes through a diffraction grating and turns into a set of IR dots. The relative geometry between the IR projector and the IR camera as well as the projected IR dot pattern are known. If we can match a dot observed in an image with a dot in the projector pattern, we can reconstruct it in 3D using triangulation. Because the dot pattern is relatively random, the matching between the IR image and the projector pattern can be done in a straightforward way by comparing small neighborhoods using, for example, normalized cross correlation. The depth value is encoded with gray values; the darker a pixel, the closer the point is to the camera in space. The black pixels indicate that no depth values are available for those pixels. This might happen if the points are too far (and the depth values cannot be computed accurately), are too close (there is a blind region due to limited fields of view for the projector and the camera), are in the cast shadow of the projector (there are no IR dots), or reflect poor IR lights </p> <p><em>HD Images</em></p> <p><br> The HD images were acquired using 5 Megapixel HD camera mounted on a mobile chassis made by Dawn Robotics, that allowed the camera to be moved freely on the scene. Both camera and mobile chassis was controlled by a Raspberry PI unit which was also responsible to position the camera in accord to the data being collected by other sources. <br> </p> <p><strong>Data</strong></p> <p>The dataset consists of three scenes consisting of various objects -- minerals, fruit, wood plastic and metal -- placed on a stand. The objects, depending on the view are partially covered and seen from different perspective. Each scene is captured from 8 different angles.</p> <p>Scene 1 (denoted <em>SceneEagle</em>) uses mostly inorganic materials, such as wood, metal, plastic and glass all placed on the vertical stand.<br> Scene 2 (<em>SceneFruit</em>) uses fruits normal and artificial, that are similar on HD photography and 3D cloud of point, but differs in hyperspectral image.<br> Scene 3 (<em>SceneFruit2</em>) uses the fruits but also includes printed full colour images of same fruits that are 2-dimensional.</p> <p> </p> <p>The data are formatted as follows:<br> - The HIS images are available in both \text{*.hdr} and \text{*.cube} formats. The separate files with calibrating panel is provided for each frame.<br> - Kinect clouds are provided in \text{*.obj} format, typical for Kinect output files.<br> - Matched Hyperspectral clouds are also provided as \text{*.obj} files<br> - HD photo files are provided in \text{*.jpg} files.<br> </p> <p><br> <strong>Acknowledgements</strong></p> <p>This work has been supported by the National Science Centre, based on decision no. DEC2012/07/N/ST6/03656.</p> <p> </p>
Dataset for: Image-based screen capturing misfolding status of Niemann-Pick type C1 identifies potential candidates for chaperone drugs
<p>Niemann-Pick disease type C is a rare, fatal neurodegenerative disorder characterized by massive intracellular accumulation of cholesterol. In most cases, loss-of-function mutations in NPC1 gene that encodes for a lysosomal cholesterol transporter NPC1 are responsible for the disease, and more than half of the mutations are considered to interfere with biogenesis or folding of the protein. Previously we have identified a series of oxysterol derivatives and phenanthridine-6-one derivatives as pharmacological chaperones, small molecules that rescue folding defective phenotypes of a mutated NPC1, and opened an avenue to develop chaperone therapy for Niemann-Pick disease type C. Here, we established an improved image-based screen for NPC1 chaperones and performed drug-repurposing screening to identify some azole antifungals, including itraconazole and posaconazole, and a kinase inhibitor lapatinib as probable pharmacological chaperones. Photo-crosslinking probes of the compounds allowed us to detect direct binding of itraconazole to a representative folding-defective mutant, NPC1-I1061T. Competitive photo-crosslinking experiments suggested that oxysterol-based chaperones and itraconazole share the same or nearby binding site(s), and sensitivity of the crosslinking to P691S mutation on the sterol-sensing domain supported currently proposed hypothesis that their binding sites are located near the domain. Although the azoles were less effective in reducing cholesterol accumulation than the oxysterol-derived chaperone or an HDAC inhibitor LBH-589, our findings should offer new starting points for developing better pharmacological chaperones for NPC1 through medicinal chemistry efforts.</p>
Dataset: Object condensation: one-stage grid-free multi-object reconstruction in physics detectors, graph, and image data
<p>This dataset is the dataset used to train and test the object condensation particle flow approach described in <a href="https://arxiv.org/abs/2002.03605">arxiv:2002.03605</a>.</p> <p>The data can be read with DeepJetCore 3.1 (https://github.com/DL4Jets/DeepJetCore)<br> The entries in the truth array are of dimension (batch, 200, N_truth). The truth inputs are:</p> <p>isElectron,<br> isGamma,<br> isPositron,<br> true_energy,<br> true_x,<br> true_y</p> <p>The entries in the feature array are of dimension (batch, 200, N_features), with the features being:</p> <p>rechit_energy,<br> rechit_x,<br> rechit_y,<br> rechit_z,<br> rechit_layer,<br> rechit_detid</p> <p>The "train.zip" file contains the training sample<br> The "test.zip" file the test sample</p> <p>The main test sample is identical to the training sample in composition, but statistically independent.<br> Other samples can be found in subfolders:</p> <p>test/flatNpart: sample with flat distribution of additional particles in the event w.r.t. each individual particle<br> Test/hiNPart: sample with up to 15 particles per event</p>
Girasol, a sky imaging and global solar irradiance dataset
<p>The energy available in Micro Grid (MG) that is powered by solar energy is tightly related to the weather conditions in the moment of generation. Very short-term forecast of solar irradiance provides the MG with the capability of automatically controlling the dispatch of energy. We propose to achieve this using a data acquisition systems (DAQ) that simultaneously records sky imaging and Global Solar Irradiance (GSI) measurements, with the objective of extracting features from clouds and use them to forecast the power produced by a Photovoltaic (PV) system. The DAQ system is nicknamed as the <em>Girasol Machine</em> (Girasol means Sunflower in Spanish). The sky imaging system consists of a longwave infrared (IR) camera and a visible (VI) light camera with a fisheye lens attached to it. The cameras are installed inside a weatherproof enclosure that it is mounted on an outdoor tracker. The tracker updates its pan an tilt every second using a solar position algorithm to maintain the Sun in the center of the IR and VI images. A pyranometer is situated on a horizontal support next to the DAQ system to measure GSI. The dataset, composed of IR images, VI images, GSI measurements, and the Sun's positions, has been tagged with timestamps.</p>
Grassland African Road Images (GARI): A Driving Dataset from Kenyan Highways and National Parks
<p>A driving dataset collected on grassland roads in Kenyan national parks and highways in western Kenya. The dataset contains images from 20 hours of driving as well the driving signals from the vehicle such as speed, steering, acceleration etc. This is the first driving dataset from Africa to the best of our knowledge. See <em>GARI_dataset_description.pdf</em> for more information.</p>
Dataset related to article "Evaluation of cell metabolic adaptation in wound and tumour by fluorescence Lifetime imaging Microscopy"
<p>This record contains data related to article "Evaluation of cell metabolic adaptation in wound and tumour by fluorescence Lifetime imaging Microscopy"</p> <p>Abstract</p> <p>Acidic pH occurs in acute wounds progressing to healing as consequence of a cell metabolic adaptation in response to injury-induced tissue hypoperfusion. In tumours, high metabolic rate leads to acidosis affecting cancer progression. Acidic pH affects activities of remodelling cells in vitro. The pH measurement predicts healing in pathological wounds and success of surgical treatment of burns and chronic ulcers. However, current methods are limited to skin surface or based on detection of fluorescence intensity of specific sensitive probes that suffer of microenvironment factors. Herein, we ascertained relevance in vivo of cell metabolic adaptation in skin repair by interfering with anaerobic glycolysis. Moreover, a custom-designed skin imaging chamber, 2-Photon microscopy (2PM), fluorescence lifetime imaging (FLIM) and data mapping analyses were used to correlate maps of glycolytic activity in vivo as measurement of NADH intrinsic lifetime with areas of hypoxia and acidification in models of skin injury and cancer. The method was challenged by measuring the NADH profile by interfering with anaerobic glycolysis and oxidative phosphorylation in the mitochondrial respiratory chain. Therefore, intravital NADH FLIM represents a tool for investigating cell metabolic adaptation occurring in wounds, as well as the relationship between cell metabolism and cancer.</p>
Tree Species Dataset consisting of Images of the Bark, Leaves or Needles
<p>This dataset consists of 3 subsets:</p> <ul> <li>Leaves of the most common Austrian broad leaf trees: Ash (25), Beech (30), Hornbeam (34), Mountain oak (22), Sycamore maple (23)</li> <li>Bark of the most common Austrian trees: Ash (34), Beech (16), Black pine (166), Fir (127), Hornbeam (42), Larch (200), Mountain oak (77), Scots pine (190), Spruce (213), Swiss stone pine (96), Sycamore maple (22)</li> <li>Needles of the most common Austrian conifers: Black pine (107), Fir (10), Larch (114), Scots pine (10), Spruce (13), Swiss stone pine (21)</li> </ul> <p>The leaf dataset consists of 134 images of five Austrian broad leaf trees which were scaled to either 800 pixel height or 600 pixel width. Every class has 25 to 34 images. While the beech, hornbeam, mountain oak and sycamore maple are complete leaves, the ash is compound, more precise a pinnate leaf.</p> <p>The dataset of bark images contains 1183 images of eleven Austrian trees. Every class has 16 to 213 images. These images were also scaled to a size of either 800 pixel height or 600 pixel width. The dataset of the black pine, fir, larch, scots pine and spruce are divided in 3 sub-classes. The first containing images of the trees when they are younger than 60, in the second one images of trees with an age of 60 to 80, and the last one with images of trees which are older than 80. These separation is necessary because especially the bark of these trees differs at different ages.</p> <p>The dataset of the needle images contains 275 of 6 Austrian conifers. Each class contains 10 to 114 images. Conifers can be divided into two classes: The first class are the fir and the spruce on which the needles grow separate on the branch and the second class are the species on which the needles grow in clusters. It can be seen that the fir, scots pine, and spruce images have been made with perfect lighting conditions, whereas the other images have been photographed in the nature.</p> <p>These datasets were gathered by employees of the ”Osterreichische ¨ Bundesforste AG“ in autumn 2009 and spring 2010.</p> <p>This database may be used for non-commercial research purpose only. If you publish material based on this database, we request you to include a reference to:</p> <p>Fiel, S. & Sablatnig, R. (2010): <em>Leaf classification using local features</em> In: Proc. of 34th annual Workshop of the Austrian Association for Pattern Recognition (AAPR), 2010, 69-74 <a href="https://cvl.tuwien.ac.at/wp-content/uploads/2014/12/fiel-oeagm10.pdf">pdf</a></p> <p>Fiel, S. (2010): <em>Automated Identification of Tree Species from Images of the Bark, Leaves or Needles,</em> Master Thesis, Vienna University of Technology <a href="https://cvl.tuwien.ac.at/wp-content/uploads/2014/12/tr32.pdf">pdf</a></p> <p>Version 2: added Bark.zip which is the selection of Bark images used in the paper.</p>
Dataset for Black Tea Fermentation Detection based on Image Processing and Machine Learning Techniques
<p>This is a dataset on black tea fermentation. The dataset contains black tea fermentation conditions and images. The fermentation conditions captured are: temperature, humidty and time. The images belong to black tea as they underwent the fermentation process. The dataset was collected in Sisibo tea factory, Kenya in July and August 2020.</p>
Amazon and Atlantic Forest image datasets for semantic segmentation
<p>This database contains images from<strong> Amazon </strong>and <strong>Atlantic Forest </strong>brazilian biomes used for training a fully convolutional neural network for the semantic segmentation of forested areas in images from the Sentinel-2 Level 2A Satellite.</p> <p>The images refer to the composition of bands 4, 3, 2 and 8. Each band was converted to a byte type (0-255).</p> <p>The images are still divided into three main sets: training, validation and testing:</p> <ol> <li><strong>Training dataset: </strong>it contains 499 and 485 GeoTIFF images (Amazon and Atlantic Forest, respectively) with 512x512 pixels and associated PNG masks (forest indicated in white and background in black color).</li> <li><strong>Validation dataset</strong>: it contains 100 GeoTIFF images for each biome with 512x512 pixels and associated PNG masks used for validation step.</li> <li><strong>Test dataset: </strong>it contains 20 GeoTIFF images for each biome with 512x512 pixels for testing.</li> </ol>
Structure Assisted Compressed Sensing Reconstruction of Undersampled AFM Images Dataset 2
<p>This deposition contains the results from a simulation of reconstructions of undersampled atomic force microscopy (AFM) images. The reconstructions were obtained using weighted iterative thresholding compressed sensing algorithms.</p> <p>The deposition consists of:</p> <ol> <li>An HDF5 database containing the results from simulations of reconstructions of undersampled atomic force microscopy images (<em>weighted_it_reconstructions.hdf5</em>).</li> <li>The Python script which was used to create the database (<em>weighted_it_reconstructions.py</em>).</li> <li>MD5 and SHA256 checksums of the database and Python script files (<em>weighted_it_reconstructions.MD5SUMS / weighted_it_reconstructions.SHA256SUMS</em>).</li> </ol> <p>The HDF5 database is licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/) . Since the CC BY 4.0 license is not well suited for source code, the Python script is licensed under the BSD 2-Clause license (http://opensource.org/licenses/BSD-2-Clause) .</p> <p><strong>The files are provided as-is with no warranty as detailed in the above mentioned licenses.</strong></p> <p>The database is split into ten parts:</p> <ol> <li>weighted_it_reconstructions.hdf5.tar.xz.part-00</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-01</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-02</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-03</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-04</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-05</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-06</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-07</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-08</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-09</li> </ol> <p>These tem parts must be concatenated before the database can be extracted from the tar.xz archive. On Unix-like systems this may be done using:</p> <p><em>$ cat weighted_it_reconstructions.hdf5.tar.xz.part-* > weighted_it_reconstructions.hdf5.tar.xz</em></p> <p>after which the archive may be extracted, e.g., using:</p> <p><em>$ tar xfJ weighted_it_reconstructions.hdf5.tar.xz</em></p> <p><strong>WARNING: The extracted HDF5 database has a size of 114 GiB.</strong></p> <p>The simulation results in the database are based on "Atomic Force Microscopy Images of Cell Specimens" and "Atomic Force Microscopy Images of Various Specimens" by Christian Rankl licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). The original images are available at http://dx.doi.org/10.5281/zenodo.17573 and http://dx.doi.org/10.5281/zenodo.60434. The original images are provided as-is without warranty of any kind. Both the original images as well as adapted images are part of the dataset. </p>
Structure Assisted Compressed Sensing Reconstruction of Undersampled AFM Images Dataset
<p>This deposition contains the results from a simulation of reconstructions of undersampled atomic force microscopy (AFM) images. The reconstructions were obtained using weighted iterative thresholding compressed sensing algorithms.</p> <p>The deposition consists of:</p> <ol> <li>An HDF5 database containing the results from simulations of reconstructions of undersampled atomic force microscopy images (<em>weighted_it_reconstructions.hdf5</em>).</li> <li>The Python script which was used to create the database (<em>weighted_it_reconstructions.py</em>).</li> <li>MD5 and SHA256 checksums of the database and Python script files (<em>weighted_it_reconstructions.MD5SUMS / weighted_it_reconstructions.SHA256SUMS</em>).</li> </ol> <p>The HDF5 database is licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/) . Since the CC BY 4.0 license is not well suited for source code, the Python script is licensed under the BSD 2-Clause license (http://opensource.org/licenses/BSD-2-Clause) .</p> <p><strong>The files are provided as-is with no warranty as detailed in the above mentioned licenses.</strong></p> <p>The database is split into four parts:</p> <ol> <li>weighted_it_reconstructions.hdf5.tar.xz.part-00</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-01</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-02</li> <li>weighted_it_reconstructions.hdf5.tar.xz.part-03</li> </ol> <p>These four parts must be concatenated before the database can be extracted from the tar.xz archive. On Unix-like systems this may be done using:</p> <p><em>cat weighted_it_reconstructions.hdf5.tar.xz.part-* > weighted_it_reconstructions.hdf5.tar.xz</em></p> <p>after which the archive may be extracted, e.g., using:</p> <p><em>tar xfJ weighted_it_reconstructions.hdf5.tar.xz</em></p> <p><strong>WARNING: The extracted HDF5 database has a size of 70 GiB.</strong></p> <p>The simulation results in the database are based on "Atomic Force Microscopy Images of Cell Specimens" by Christian Rankl licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). The original images are available at http://dx.doi.org/10.5281/zenodo.17573. The original images are provided as-is without warranty of any kind. Both the original images as well as adapted images are part of the dataset. </p>
Human Interaction Image (HII) dataset
<p>The Human Interaction Image (HII) dataset is a new dataset containing Web images from Commercial Search Engines (Google, Bing and Flickr). We use keyword search to collect images corresponding to four types of interactions: handshake, highfive, hug, kiss. Then we manually filter the irrelevant images. The dataset contains 2410 images with at least 550 images per interaction.</p> <p>The dataset can be applied, but not limited to the following research areas:</p> <ul> <li>interaction recognition/prediction</li> <li>action recognition</li> <li>video analysis</li> <li>transfer learning</li> </ul> <p>Please cite the following paper if you use the HII dataset in your work (papers, articles, reports, books, software, etc):</p> <ul> <li>J. Li, Y. Wong, Q.Zhao, M. Kankanhalli<br> <strong>Attention Transfer from Web Images for Video Recognition</strong><br> <em>ACM Multimedia</em>, 2017.<br> http://doi.org/10.1145/3123266.3123432</li> </ul>
Datasets, models and demos associated to "Celldetective: an AI-enhanced image analysis tool for unraveling dynamic cell interactions"
<p>This repository contains datasets, models and demos associated to <a href="https://github.com/remyeltorro/celldetective">Celldetective</a>, a software for single-cell analysis from multimodal time lapse microscopy images. </p> <h1>Demos</h1> <h2>Cell-cell interaction assay: ADCC</h2> <p>We imaged a co-culture of MCF-7 breast cancer cells (targets) and human primary NK cells (effectors), interacting in the presence of bispecific antibodies, to measure antibody dependent cellular cytotoxicity (ADCC). The nuclei of all cells are marked with the Hoechst nuclear stain, the dead nuclei with the propidium iodide nuclear stain, the cytoplasm of the NK cells with CFSE. The system in epifluorescence and brightfield at either 20 or 40X magnification. We provide a single position demo for the ADCC assay, as "demo_adcc.zip". After unzipping, the demo_adcc folder can be loaded in Celldetective for testing. </p> <h2>Cell-surface interaction assay: RICM</h2> <p>We imaged human primary NK cells engaging in spreading with a surface coated with a bispecific antibody similar to the one used in the ADCC assay (replacing the target cells with a flat surface). The system is imaged using the RICM technique. Images are normalized using a median estimate of the background, pooled from all the positions in a well and dividing the images by this estimate. Here, we provide a single position demo for the cell-surface interactiona assay imaged in RICM, as "demo_ricm.zip". As above, after unzipping, the experiment can be tested and processed in Celldetective.</p> <h1>Datasets</h1> <h2>Image annotations for segmentation</h2> <h3>Cell-cell interaction assay: ADCC</h3> <p>We generated two sets of annotations from images of a co-culture of MCF-7 breast cancer cells and human primary NK cells, interacting in the presence of bispecific antibodies, to measure antibody dependent cellular cytotoxicity (ADCC). Since there are two separate cell populations of interest, the targets (MCF-7) and effectors (NK cells), we curated two datasets. Each sample in a dataset consists of a multichannel image (up to five channels in the context of ADCC, among brightfield , Hoechst nuclear stain, PI nuclear stain, CFSE, LAMP1), the associated instance segmentation annotation for the population of interest and a json file summarizing the content of each channel and the spatial calibration of the image. These sample data are generated directly in Celldetective, using a custom napari plugin.</p> <ul> <li>db_mcf7_nuclei_w_lymphocytes: MCF-7 cell nuclei are annotated specifically on images where primary NK cells (or rarely primary T cells), and RBCs co-exist. The annotation exploits up to four channels simultaneously.</li> <li>db_primary_NK_w_mcf7: human primary NK cells, with annotated cytoplasm (mostly from CFSE) but exploiting brightfield and Hoechst to segment out of focus or poorly labelled cells.</li> </ul> <p>These datasets are used to train several segmentation models to segment on one hand the MCF-7 nuclei and on the other hand the primary NK cells.</p> <h3>Cell-surface interaction assay: RICM</h3> <ul> <li>db_spreading_lymphocytes: we provide a dataset of primary NK cells (and occasionnaly mice T cells) imaged in RICM (with sometimes paired brightfield images). Cells are detected as soon as they start forming interferences on the image (hovering behavior). A pre-annotation was performed using a threshold based segmentation on the RICM modality. Manuel separation of cell-cell contacts and removal of false positive objects was performed by an expert annotator (using brightfield when available). RBCs are ignored in the annotations. </li> </ul> <h2>Single-cell signal annotations for classification and regression</h2> <h3>Cell-cell interaction assay: ADCC</h3> <p>We generated several signal classification/regression datasets with Celldetective to characterize the ADCC assay. Briefly, for a given event cells can be classified as "the event occured during the observation", "no event occured during the observation", "the event already occured prior to observation". If the event occurred during the observation, we can estimate when (the regression). Each single-cell is a dictionary with a collection of signals. The attribute "class" sets the class and "t0" the time of event (default is -1 for absence of event). </p> <ul> <li>db-si-NucPI: classification and regression of single-cells with respect to lysis events characterized by a strong PI increase upon lysis (also associated with decreasing nuclear area and sometimes a decreasing Hoechst)</li> <li>db-si-NucCondensation: classification and regression of single-cells with respect to nucleus shrinking events characterized by a decreasing nuclear area (UPDATE on 23/01/2024)</li> </ul> <h1>Models</h1> <h2>Segmentation models</h2> <h3>Generalist models</h3> <p>We integrated in Celldetective select published models for cellular segmentation from StarDist and Cellpose. We wraped the models with an input configuration to help Celldetective handle the normalization, rescaling and channel selection upon inference. </p> <ul> <li>Cellpose [1,2]: <em>cyto3</em>, <em>livecell</em>, <em>tissuenet</em>, <em>nuclei</em></li> <li>StarDist [3]: <em>versatile_fluo</em>, <em>versatile_he</em></li> </ul> <p>If you use any of these models your research, don't forget to cite the StarDist or Cellpose papers accordingly!</p> <h3>ADCC models</h3> <ul> <li>MCF-7 (in the presence of lymphocytes): <em>mcf7_nuc_multimodal, mcf7_nuc_stardist_transfer</em></li> <li>primary NKs (in the presence of MCF-7): <em>primNK_multimodal</em>, <em>primNK_SD</em>, <em>primNK_cfse</em></li> </ul> <h3>Spreading-assay models</h3> <ul> <li>Lymphocytes: <em>lymphocytes_ricm</em></li> </ul> <h2>Signal analysis models</h2> <p>We developed Deep Learning models that classify and regress the time of events from single-cell signals, applied to the ADCC assay.</p> <ul> <li> lysis detection: <em>lysis_H_PI</em>, <em>lysis_PI_area</em><em>. </em>Detect lysis events characterized at least by an increase of PI from one or more measurements (respectively PI+Hoechst and PI+nucleus area, trained on db-si-NucPI)</li> <li>nucleus shrinking detection:<em> NucCond</em>. Detect nucleus shrinking events from nuclear area signal (db-si-NucCondensation)</li> </ul> <h1>References</h1> <ol> <li>Stringer, C., Wang, T., Michaelos, M. & Pachitariu, M. Cellpose: a generalist algorithm for cellular segmentation. Nat Methods 18, 100–106 (2021).</li> <li>Pachitariu, M. & Stringer, C. Cellpose 2.0: how to train your own model. Nat Methods 19, 1634–1641 (2022).</li> <li>Schmidt, U., Weigert, M., Broaddus, C. & Myers, G. Cell Detection with Star-Convex Polygons. in Medical Image Computing and Computer Assisted Intervention – MICCAI 2018 (eds. Frangi, A. F., Schnabel, J. A., Davatzikos, C., Alberola-López, C. & Fichtinger, G.) 265–273 (Springer International Publishing, Cham, 2018). doi:10.1007/978-3-030-00934-2_30.</li> </ol> <p> </p> <p> </p>
Five-fold training dataset of fossil pollen images from Burgäschisee used for automated fossil pollen identification (von Allmen et al. study)
<p>This dataset consists of a training dataset for the CNN model containing pollen grain images of nine common pollen taxa, one marker class (Lycopodium clavatum) and four abundant non-pollen debris classes. The dataset was split five-times so that each image is part of the validation dataset in just one of these splits. Additionally, the dataset contains annotated images used to train the object detection model and a dataset of images that were used to evaluate the performance of the CNN model. For further information on the datsets themselves and how they were used the reader may refer to the github repository here attached (<a href="https://github.com/RobinVonAllmen/FOSSILPOLLEN">https://github.com/RobinVonAllmen/FOSSILPOLLEN)</a></p>
Pix2pix training dataset for predicting nuclei form brightfield images
<p>Dataset for training pix2pix model to predict nuclei from brightfield images of MDA-MB-231. </p><p>This dataset contains 1600 brightfield and 1600 nuclei images. </p><p>The nuclei were labeled using sir-DNA (Cytoskeleton). Cells were imaged live (37 °C, 5% CO2) using a Nikon Eclipse Ti2-E microscope (Nikon) equipped with an sCMOS Orca Flash4.0 camera (Hamamatsu) and controlled by the NIS-Elements software (Nikon, v 5.11.01). MDA-MB-231 cells were imaged using a 20× Nikon CFI Plan Apo Lambda objective (NA 0.75), either 1 frame per minute for 2 hours or 1 frame every 5 minutes for 17 hours. In these experiments, a camera binning of 2 × 2 was used. </p>
Teleseismic reverse time migration image dataset of southwest Japan in "Three-dimensional teleseismic elastic reverse-time migration with deconvolution imaging condition and its application to southwest Japan"
<p>This dataset contains the teleseismic elastic reverse time migration results of southwest Japan used in the manuscript entitled "Three-dimensional teleseismic elastic reverse-time migration with deconvolution imaging condition and its application to southwest Japan". submitted to Journal of Geophysical Research Letters.</p>
TINKER_WP3_2D&3D images - profile scans dataset_221123
<p>2D and 3D images of PCBs showing the gap 2D and 3D information. Moreover, profile measurements are also included in x and y axes.</p>
Rulers2023: An Annotated Dataset of Synthetic and Real Images for Ruler Detection Using Deep Learning
<p>Annotated datasets of synthetic and real ruler images:<br>1. Synthetic-train<br>2. Real-train<br>3. Real-test</p><p><strong>Citation:</strong> Matuzevičius D. Rulers2023: An Annotated Dataset of Synthetic and Real Images for Ruler Detection Using Deep Learning. <i>Electronics</i>. 2023; 12(24):4924. https://doi.org/10.3390/electronics12244924</p><p> </p>
A large and diverse brain organoid dataset of 1,400 cross-laboratory images of 64 trackable brain organoids from four different clones
<p>This dataset is presented in the paper <strong><span>A large and diverse brain organoid dataset of 1,400 cross-laboratory images of 64 trackable brain organoids from four different clones</span></strong></p> <p> </p> <p>This dataset encompasses two sources of data:</p> <ol> <li>A comma-separated values ('CSV') file. This file serves as a key to our dataset with one image per row. Each image is represented by its image identifier ('img_id') with the format [org_id]_[clone]_d[imaging_day]_[imaging_lab]. For each image, the CSV file also specifies the organoid size for convenience. Alternatively, the organoid size can be calculated using the ground truth organoid segmentation (org_segGT). </li> <li>For each row of the CSV file, we provide the image and org_segGT. For Lab A, the images are in JPEG format. For lab B, the images are in TIF format. Org_segGT is a manually created binary 2D NumPy array with the same size as the image (1024 x 768 for lab A, 1388 x 1040 for lab B). A value of 1 in org_segGT at position (x, y) means that the same position (x, y) in the corresponding image is covered by the organoid. The image file and the org_segGT file have the following format: [img_id].[jpg|tif] and [img_id].npy. For day 12, organoids were imaged before and after embedding from 96-well plates in 12-well plates, allowing the investigation of well-specific optical properties.</li> </ol> <p>For segmentation and growth monitoring using this dataset, please see <a href="https://github.com/deiluca/robust_monitoring_organoid_growth">https://github.com/deiluca/robust_monitoring_organoid_growth</a>.</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.