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.
57
datasets available to search
ShareScore release 0.9.0
Dataset results
57 results for “Chest X-ray”
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>
From Generalist to Specialist: Incorporating Domain-Knowledge into Flamingo for Chest X-Ray Report Generation
<p>This subset of the MIMIC-CXR split file contains the study identifiers and paths to the chest X-ray images used for training, validation and testing of all models presented in the paper: "From Generalist to Specialist: Incorporating Domain-Knowledge into Flamingo for Chest X-Ray Report Generation".</p>
Multienergy Fan Beam Computed Tomography Dataset of a Bird Chest Imaged with 3 Different X-ray Spectra
<p><strong>Summary</strong></p> <p>This dataset is a collection of X-ray projection data of a biological imaging phantom (a bird chest) imaged in an X-ray microtomography scanner, using three different X-ray spectra. The dataset also includes a metadata file for each of the scans, specifying the scan geometry and other important scan parameters, as well as photographs and example reconstructions. The dataset is designed for use in algorithm development for multienergy computed tomography.</p> <p> </p> <p><strong>Description</strong></p> <p><em>Sample Information</em></p> <p>The sample is the chest of a common quail (<em>Coturnix coturnix</em>) bird obtained frozen from a local supermarket. The chest section of the frozen bird was removed using a handsaw, and left to melt and settle in a sample holder before imaging.</p> <p><em>Scanner</em></p> <p>The measurement data were acquired using an X-ray microtomography scanner in the University of Helsinki Micro-CT Laboratory. The scanner uses cone beam geometry and it is equipped with an end-window tube with a tungsten target.</p> <p><em>Scan Settings</em></p> <p>The dataset consists of three consecutive scans made using identical geometry but different X-ray spectra and detector exposure times. For each scan, 720 X-ray projections were acquired using an angle increment of 0.5 degrees. Multiple frames were averaged for each projection in order to increase signal-to-noise ratio. The scan geometry and the energy-specific settings are summarized in the following two tables.</p> <p><strong>Table 1.</strong> Imaging geometry used for collecting the data.</p> <table> <tbody> <tr> <td><strong>Parameter</strong></td> <td><strong>Value</strong></td> </tr> <tr> <td>Focus-center distance</td> <td>252 mm</td> </tr> <tr> <td>Focus-detector distance</td> <td>420 mm</td> </tr> <tr> <td>Geometric magnification</td> <td>5/2</td> </tr> <tr> <td>Detector pixel size</td> <td>0.200 mm</td> </tr> <tr> <td>Effective pixel size</td> <td>0.120 mm</td> </tr> <tr> <td>Projection size</td> <td>552 x 576 pixels</td> </tr> <tr> <td>Angular range</td> <td>360'</td> </tr> <tr> <td>#projections</td> <td>720</td> </tr> </tbody> </table> <p><strong>Table 2.</strong> Energy-specific settings used for collecting the data.</p> <table> <tbody> <tr> <td>Energy label</td> <td><em>U</em> (kV)</td> <td>Filtration</td> <td><em>I</em> (μA)</td> <td>Exposure time (ms)</td> <td>Frame averaging</td> </tr> <tr> <td><em>E1</em></td> <td>50</td> <td>None</td> <td>300</td> <td>125</td> <td>4</td> </tr> <tr> <td><em>E2</em></td> <td>80</td> <td>1 mm Al</td> <td>180</td> <td>125</td> <td>4</td> </tr> <tr> <td><em>E3</em></td> <td>120</td> <td>0.5 mm Cu</td> <td>120</td> <td>250</td> <td>4</td> </tr> </tbody> </table> <p><em>Data Post-Processing</em></p> <p>Before the scans were made, a dark current image and flat-field image were acquired for each scan setting. During the scans, dark current subtraction and flat-field correction were automatically applied to the X-ray projections by the measurement software.</p> <p><em>Data Contents</em></p> <p>This dataset contains the following files:</p> <ul> <li>The raw projection data (.tif format) for each scan and a metadata file (.txt format) describing the measurement setup, with formatting that is both human-readable and machine-readable.</li> <li>Pre-created 2D sinograms for each energy level. The sinograms have been created from the central plane of the cone beam, which reduces to fan beam geometry. The sinograms are stored in Matlab's .mat file format in data structures which also contain metadata on the measurement.</li> <li>Photographs taken during the measurement process.</li> <li>Example filtered backprojection (FBP) reconstructions of the central plane of the phantom for each energy. The reconstructions were computed using the Phoenix datos|x CT software provided with the microtomography scanner</li> </ul> <p> </p> <p><strong>Research Group</strong></p> <p>This dataset was produced by the Inverse Problems research group at the Department of Mathematics and Statistics at the University of Helsinki, Finland (<a href="https://www.helsinki.fi/en/researchgroups/inverse-problems">https://www.helsinki.fi/en/researchgroups/inverse-problems</a>) in collaboration with the Computational Physics and Inverse Problems research group at the University of Eastern Finland, Finland (<a href="https://sites.uef.fi/inverse">https://sites.uef.fi/inverse</a>) and the X-ray Laboratory at the Department of Physics at the University of Helsinki, Finland (<a href="https://www.helsinki.fi/en/researchgroups/x-ray-laboratory">https://www.helsinki.fi/en/researchgroups/x-ray-laboratory</a>).</p> <p> </p> <p><strong>Previous Use</strong></p> <p>This dataset has been used in the following publications:</p> <p>Jussi Toivanen, Alexander Meaney, Samuli Siltanen, Ville Kolehmainen. Joint reconstruction in low dose multi-energy CT. <em>Inverse Problems and Imaging</em>, 2020, 14(4): 607-629. doi: <a href="https://doi.org/10.3934/ipi.2020028" target="_blank" rel="noopener">10.3934/ipi.2020028</a>.</p> <p>E. Cueva, A. Meaney, S. Siltanen, M. J. Ehrhardt. Synergistic multi-spectral CT reconstruction with directional total variation. <em>Philos Trans A Math Phys Eng Sci</em>. 2021 Aug 23;379(2204):20200198. doi: <a href="https://doi.org/10.1098/rsta.2020.0198">10.1098/rsta.2020.0198</a>.</p> <p> </p> <p><strong>Additional Links</strong></p> <p>To get started with the data, we recommend looking at the HelTomo toolbox, specifically created for working with CBCT data collected by the Inverse Problems research group, and available at <a href="https://se.mathworks.com/matlabcentral/fileexchange/74417-heltomo-helsinki-tomography-toolbox">https://se.mathworks.com/matlabcentral/fileexchange/74417-heltomo-helsinki-tomography-toolbox</a>.</p> <p> </p> <p><strong>Acknowledgements</strong></p> <p>We wish to thank laboratory engineer Heikki Suhonen for his guidance and assistance in conducting the measurements.</p> <p> </p> <p><strong>Contact Details</strong></p> <p>For more information or guidance in using these datasets, please contact alexander.meaney [at] helsinki.fi.</p>
dataset- Tuberculosis detection using Squid Game Optimization with Deep Learning Model on Chest X-Ray Images
Open the record for dataset details and reuse information.
Iterative evaluation of mobile computer-assisted digital chest x-ray screening for TB improves efficiency, yield, and outcomes in Nigeria
<p>Wellness on Wheels (WoW) is a model of mobile systematic tuberculosis (TB) screening of high-risk populations combining digital chest radiography with computer-aided automated detection (CAD) and chronic cough screening to identify presumptive TB clients in communities, health facilities, and prisons in Nigeria. The model evolves to address technical, political, and sustainability challenges.</p> <p>Screening methods were iteratively refined to balance TB yield and feasibility across heterogeneous populations. Performance metrics were compared over time. Screening volumes, risk mix, number needed to screen (NNS), number needed to test (NNT), sample loss, TB treatment initiation and outcomes. Efforts to mitigate losses along the diagnostic cascade were tracked. Participants with high likelihood on CAD4TB (≥80) who tested negative on a single spot GeneXpert were followed-up to assess TB status at six months.</p> <p>An experimental calibration method achieved a viable CAD threshold for testing. High-risk groups and key stakeholders were engaged. Operations evolved in real-time to fix problems. Incremental improvements in mean client volumes (128 to 140/day), target group inclusion (92% to 93%), on-site testing (84% to 86%), TB treatment initiation (87% to 91%), and TB treatment success (71% to 85%). Attention to those as highest risk boosted efficiency (the NNT declined from 8.2 ± SD8.2 to 7.6 ± SD7.7). Clinical diagnosis was added after follow-up among those with ≥ 80 CAD scores initially spot-sputum negative found 11 additional TB cases (6.3%) after 121 person-years of follow-up.</p> <p>Iterative adaptation in response to performance metrics foster feasible, acceptable, and efficient TB case-finding in Nigeria. High CAD scores can identify subclinical TB and those at risk of progression to bacteriologically-confirmed TB disease in the near term.</p> <p>Policy makers, donors, and community advocates are hesitant to invest in the steep infrastructure costs for mobile digital chest x-ray and GeneXpert MTB/RIF (dCXR/GXP) laboratories without a better understanding of how to maximize and sustain their impact. It is rarely possible to conduct the months of local CAD calibration recommended by experts via costly universal testing with a reference standard.4,9 Stakeholder needs and resource limitations require a more rapid and cost-conscious means of setting a sustainable algorithm. Viable, field-robust methodologies are needed, and optimization strategies informed by routine field findings were lacking. A precise assessment of the contribution of routine mobile TB screening has been challenging because few authors fully disaggregate losses along the diagnostic cascade or track TB treatment outcomes. Publication bias has limited access to results of active case finding pilots with suboptimal risk group targeting, community engagement, yield, or treatment outcomes.10–14 Evaluations (and scrutiny) of routine data are needed that make the demands, constraints, costs and choices facing implementers more explicit.</p>
Artificial Intelligence and COVID-19 using chest CT scan and chest X-ray images: Machine Learning and Deep Learning Approaches for Diagnosis and Treatment
<p>We uploaded the Table of included articles in the systematic review "Artificial Intelligence and COVID-19 using chest CT scan and chest X-ray images: Machine Learning and Deep Learning Approaches for Diagnosis and Treatment"</p>
A Randomized Controlled Trial of Lung Ultrasound Compared to Chest X-ray for Diagnosing Pneumonia in the Emergency Department
ClinicalTrials.gov study NCT01654887. IPD Sharing: Not stated. Countries: 1. Publications: 5.
Iterative evaluation of mobile computer-assisted digital chest x-ray screening for TB improves efficiency, yield, and outcomes in Nigeria
Open the record for dataset details and reuse information.
VinDr Chest X-ray Datset
<p>This zip file contains train and test data of VinDr Chest X-ray images.</p>
Chest X-Ray Image Dataset: A Resource for Medical Diagnosis and Machine Learning
<p>The Chest X-Ray Image Dataset is an extensive collection designed to support medical research and the development of diagnostic tools for COVID-19 detection. It consists of two distinct classes: COVID-19 affected X-ray images and normal X-ray images of the chest area, each covering the full lungs. This dataset provides a diverse range of X-ray images, capturing the unique characteristics of both healthy and COVID-19 affected lungs, making it an invaluable resource for training and testing machine learning models in medical image classification and analysis.</p>
Comparison of Bedside Ultrasound With Chest X-ray for Confirmation of Central Venous Catheter Position
ClinicalTrials.gov study NCT02959203. IPD Sharing: UNDECIDED. Countries: 1. Publications: 16.
Multicenter Validation Study of an Artificial Intelligence Tool for Automatic Classification of Chest X-rays
ClinicalTrials.gov study NCT04991987. IPD Sharing: Not stated. Countries: 1. Publications: 6.
Clinical Application of Automated Interpretation System for Chest X-Ray Images Based on Multimodal Large Models
ClinicalTrials.gov study NCT07117266. IPD Sharing: UNDECIDED. Countries: 1. Publications: 3.
Measurement of Intravascular Volume Using USG of IVC Diameter Compared to VPW Chest X-ray
ClinicalTrials.gov study NCT03535038. IPD Sharing: UNDECIDED. Countries: 1. Publications: 38.
Effectiveness of Computer-Aided Detection Chest X-Ray Screening for Improving Tuberculosis Diagnostic Yield in Chinese Primary Health Care Settings: Study Protocol for a Prospective Cluster Randomized
ClinicalTrials.gov study NCT06963606. IPD Sharing: YES. Countries: 1. Publications: 1.
AI Assisted Detection of Chest X-Rays
ClinicalTrials.gov study NCT06075836. IPD Sharing: Not stated. Countries: 1. Publications: 12.
Deep Learning Using Chest X-Rays to Identify High Risk Patients for Lung Cancer Screening CT
ClinicalTrials.gov study NCT06910956. IPD Sharing: NO. Countries: 1. Publications: 3.
Identify Coronavirus Disease by Chest X-ray
ClinicalTrials.gov study NCT05216471. IPD Sharing: UNDECIDED. Countries: 1. Publications: 5.
Retrospective Analysis of Chest X-ray Severity Scoring System of COVID-19 Pneumonia
ClinicalTrials.gov study NCT05181696. IPD Sharing: Not stated. Countries: 1. Publications: 5.
Chest X-Ray Image Diagnosis and Report Generation Dedicated Model Based on Deepseek
ClinicalTrials.gov study NCT06874647. IPD Sharing: NO. Countries: 1. Publications: 1.
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.