Find research datasets worth reusing

opencc-by-4.0Aug 2021View details →

MEDIC: A Multi-Task Learning Dataset for Disaster Image Classification

Recent research in disaster informatics demonstrates a practical and important use case of artificial intelligence to save human lives and suffering during natural disasters based on social media contents (text and images). While notable progress has been made using texts, research on exploiting the images remains relatively under-explored. To advance image-based approaches, we propose MEDIC\footnote{Available~at: \url{https://crisisnlp.qcri.org/medic/index.html}}, which is the largest social media image classification dataset for humanitarian response consisting of 71,198 images to address four different tasks in a multi-task learning setup. This is the first dataset of its kind: social media images, disaster response, and multi-task learning research. An important property of this dataset is its high potential to facilitate research on \textit{multi-task learning}, which recently receives much interest from the machine learning community and has shown remarkable results in terms of memory, inference speed, performance, and generalization capability. Therefore, the proposed dataset is an important resource for advancing image-based disaster management and multi-task machine learning research.

opencc-by-4.0Jul 2022View details →

Imaging data of mechanically loaded, micro-patterned, silk-reinforced cellulose films with gold coating for flexible electrodes in medical implants

Neurodegenerative diseases can be treated using a functional interface between the physically soft tissue such as brain and the man-made electrodes. The orders of magnitude harder neural probes cause local injuries, due to periodic micromovements owing to breathing and pulsatile blood flow leading to encapsulation and related collapsing signals. An alternative to the currently used neural implant films including polyimide, poly(p-xylylene), SU-8 - epoxy-based negative photoresist, liquid crystal polymer, and benzocyclobutene is the natural polymer cellulose with an elastic modulus between 100 and 200 MPa. This article elucidates the measurement of the mechanical properties of bare as well as mono- and double-layer silk-reinforced cellulose in phosphate-buffered saline using a universal testing machine. In addition, the article contains electron microscopy data of these micro-structured, gold-coated films subsequent to peel-off tests to access the impact of micro-structures on gold adhesion on cellulose. These imaging data were completed by electron micrographs of mechanically loaded gold-coated cellulose films to demonstrate the impact of micro-structures on crack formation. Finally, the phosphate-buffered saline-induced swelling of the micro-structure was visualized by electron micrographs obtained before and after two-month storage in air and phosphate-buffered saline, respectively.

opencc-by-4.0May 2011View details →

Figure1. Three-layer image decomposition with content protection-Access Management in Medical Image Databases Based on New Format and Contents Protection with Inverse Pyramid Decomposition

The image preparation for the image database with layered access is shown on Fig. 1. The image is archived layer by layer and the watermarks are inserted together with the image processing. The ROI (if there is one in the image) is processed in such a way, that to permit direct access for authorized users (separate pyramid is developed for the ROI representation).

opencc-by-4.0Oct 2024View details →

Dataset for Medical Image Processing in Python Carpentries lesson

This dataset contains a collection of medical imaging files for use in the <a href="https://github.com/esciencecenter-digital-skills/medical-image-processing">"Medical Image Processing with Python" lesson</a>, originally developed by the <a href="https://www.esciencecenter.nl/">Netherlands eScience Center</a>.  The dataset includes: <ol> <li>SimpleITK compatible files: MRI T1 and CT scans (training_001_mr_T1.mha, training_001_ct.mha), digital X-ray (digital_xray.dcm in DICOM format), neuroimaging data (A1_grayT1.nrrd, A1_grayT2.nrrd). Data have been downloaded from <a href="https://insightsoftwareconsortium.github.io/SimpleITK-Notebooks/Python_html/00_Setup.html">here</a>. </li> <li>MRI data: a T2-weighted image (OBJECT_phantom_T2W_TSE_Cor_14_1.nii in NIfTI-1 format). Data have been downloaded from <a href="https://zenodo.org/records/6467772">here</a>. </li> <li>Example images for the machine learning lesson: chest X-rays (rotatechest.png, other_op.png), cardiomegaly example (cardiomegaly_cc0.png).</li> <li>Array data: Array data for the Intro to Medical Imaging lesson. Numpy arrays were created by processing and manipulation of publicly available data i.e. from <a href="https://doi.org/10.1109/TNS.1974.6499235">the Schepp Logan phantom</a> and from the <a href="https://fastmri.med.nyu.edu/">NYU FastMRI dataset</a> <div> </div> </li> <li>Additional data: to be added</li> </ol> These files represent various medical imaging modalities and formats commonly used in clinical research and practice. They are intended for educational purposes, allowing students to practice image processing techniques, machine learning applications, and statistical analysis of medical images using Python libraries such as scikit-image, pydicom, and SimpleITK.

opencc-by-4.0Jul 2023View details →

COVID-19 medical image datasets

This repository contains three curated datasets for the medical image classification described in the paper entitled "Explainable deep transfer learning fine-tunning with domain adaptation enables trustworthy COVID-19 prediction".

opencc-by-4.0Nov 2020View details →

MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis

This data repository for MedMNIST v1 is out of date! Please check the <a href="http://medmnist.github.io">latest version</a> of MedMNIST v2.    Abstract We present MedMNIST, a collection of 10 pre-processed medical open datasets. MedMNIST is standardized to perform classification tasks on lightweight 28x28 images, which requires no background knowledge. Covering the primary data modalities in medical image analysis, it is diverse on data scale (from 100 to 100,000) and tasks (binary/multi-class, ordinal regression and multi-label). MedMNIST could be used for educational purpose, rapid prototyping, multi-modal machine learning or AutoML in medical image analysis. Moreover, MedMNIST Classification Decathlon is designed to benchmark AutoML algorithms on all 10 datasets; We have compared several baseline methods, including open-source or commercial AutoML tools. The datasets, evaluation code and baseline methods for MedMNIST are publicly available at <a href="https://medmnist.github.io/">https://medmnist.github.io/</a>.   Please note that this dataset is NOT intended for clinical use.   We recommend our official <a href="https://github.com/MedMNIST/MedMNIST">code</a> to download, parse and use the MedMNIST dataset: <blockquote> <pre>pip install medmnist</pre> </blockquote>   Citation and Licenses If you find this project useful, please cite our ISBI'21 paper as:      Jiancheng Yang, Rui Shi, Bingbing Ni. "MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis," arXiv preprint arXiv:2010.14925, 2020. or using bibtex:      @article{medmnist,          title={MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis},          author={Yang, Jiancheng and Shi, Rui and Ni, Bingbing},          journal={arXiv preprint arXiv:2010.14925},          year={2020}      } Besides, please cite the corresponding paper if you use any subset of MedMNIST. Each subset uses the same license as that of the source dataset.   PathMNIST Jakob Nikolas Kather, Johannes Krisam, et al., "Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study," PLOS Medicine, vol. 16, no. 1, pp. 1–22, 01 2019. License: <a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>   ChestMNIST Xiaosong Wang, Yifan Peng, et al., "Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases," in CVPR, 2017, pp. 3462–3471. License: <a href="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0</a>   DermaMNIST Philipp Tschandl, Cliff Rosendahl, and Harald Kittler, "The ham10000 dataset, a large collection of multisource dermatoscopic images of common pigmented skin lesions," Scientific data, vol. 5, pp. 180161, 2018. Noel Codella, Veronica Rotemberg, Philipp Tschandl, M. Emre Celebi, Stephen Dusza, David Gutman, Brian Helba, Aadi Kalloo, Konstantinos Liopyris, Michael Marchetti, Harald Kittler, and Allan Halpern: “Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)”, 2018; arXiv:1902.03368. License: <a href="https://creativecommons.org/licenses/by-nc/4.0/">CC BY-NC 4.0</a>   OCTMNIST/PneumoniaMNIST Daniel S. Kermany, Michael Goldbaum, et al., "Identifying medical diagnoses and treatable diseases by image-based deep learning," Cell, vol. 172, no. 5, pp. 1122 – 1131.e9, 2018. License: <a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>   RetinaMNIST DeepDR Diabetic Retinopathy Image Dataset (DeepDRiD), "The 2nd diabetic retinopathy – grading and image quality estimation challenge," https://isbi.deepdr.org/data.html, 2020. License: <a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>   BreastMNIST Walid Al-Dhabyani, Mohammed Gomaa, Hussien Khaled, and Aly Fahmy, "Dataset of breast ultrasound images," Data in Brief, vol. 28, pp. 104863, 2020. License: <a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>   OrganMNIST_{Axial,Coronal,Sagittal} Patrick Bilic, Patrick Ferdinand Christ, et al., "The liver tumor segmentation benchmark (lits)," arXiv preprint arXiv:1901.04056, 2019. Xuanang Xu, Fugen Zhou, et al., "Efficient multiple organ localization in ct image using 3d region proposal network," IEEE Transactions on Medical Imaging, vol. 38, no. 8, pp. 1885–1898, 2019. License: <a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a> <div> <div class="gtx-trans-icon"> </div> </div>

dryad36/100

The role of health sciences libraries in supporting medical image consent standards survey documentation

Objective: To determine if health sciences library workers were familiar with best practices regarding informed consent for the publication of medical images and if they incorporate the recommendations into their professional work. Methods: A survey was developed by the authors and distributed to library listservs in the United States. The results of the survey were tabulated in R. Results: A total of 90 respondents were included in the data analysis with all respondents reporting multiple responsibilities in their professional role. While the majority of library workers (59%) were familiar with the best practices, few incorporated the recommendations into their everyday professional work. Conclusions: The professional work of health sciences library workers does not appear to include a significant inclusion of the best practices for the informed consent for the publication of medical images. There is a need for future research to better understand how library workers can better incorporate their knowledge of medical image publication consent standards into their work.

opencc-zeroApr 2024View details →

opencc-by-2.0Nov 2021View details →

Radiation Protection Perspective to Recurrent Medical Radiological Imaging

The topic of recurrent radiological imaging of patients attracted attention due to the recent studies indicating the magnitude of this phenomenon and the associated higher cumulated individual exposure to be more extensive than previously known. Recurrent imaging is used for managing various health conditions and chronic diseases such as malignancies, trauma, end-stage kidney disease, cardiovascular diseases, Crohn’s disease, urolithiasis, cystic pulmonary disease. The published studies, although available from only a part of the world, triggered discussion at international level, including two IAEA technical meetings with representatives of the IAEA Member states and international organization. The conclusions to date were reflected in the published Joint Position Statement and Call for Action by nine international organizations aimed to prompt greater dialogue and engage different stakeholders in developing and implementing strategies and solutions focused upon improved radiation protection of patients with medical conditions which prompt more frequent imaging procedures. Such actions include improved access to dose saving imaging technologies; improved imaging strategies and appropriateness process; specific optimization tailored to the clinical condition and patient habitus; wider utilization of the automatic exposure monitoring systems with an integrated option for individual exposure tracking in standardized patient-specific risk metrics; improved training and communication. These might need strengthening in the radiation protection framework to ensure that patients with medical conditions which prompt more frequent imaging procedures receive needed medical care, without undue exposure to ionizing radiation. Standardized and easily available dose information in patient-specific metrics is needed to improve risk quantification. Consensus is still lacking on the proper utilization of the dose information from the previous procedures, and the concern of misuse and misinterpretation, especially by referring physicians and patients, needs to be addressed. Like any other aspect of medical uses of ionizing radiation, the competence and awareness of users of dose information is paramount, and this is linked to the knowledge, education, training and communication. The integration of the clinical and exposure history data will support research studies and improved knowledge about radiation risks from low doses and individual radiosensitivity. The radiation protection framework will need to respond to the challenge of recurrent imaging and high individual doses. The radiation protection perspective complements the clinical perspective, and the risk to benefit analysis must account in holistic for all incidental and long-term benefits and risks for patients, their clinical history and specific needs. This is a step toward the patient-centric healthcare.

opencc-by-4.0May 2011View details →

Figure 2. Hierarchical access to the image DB-Access Management in Medical Image Databases Based on New Format and Contents Protection with Inverse Pyramid Decomposition

The structure of the hierarchical access to the image database contents is shown on Fig. 2.

opencc-by-4.0Oct 2024View details →

Large Scale Medical Image Dataset

Multiple dataset from different sources has been aggregated to create a large-scale medical image benchmark dataset in order to measure its performance. As each of the dataset’s images are of different sizes, the images are resized to 3 × 224 × 224 before the training process. This dataset contains total of 35 diseases of 4 different modality and is divided into train, validation, and test with a ratio of 7 : 1 : 2.

opencc-by-4.0Jun 2021View details →

pGAN Synthetic Dataset: A Deep Learning Approach to Private Data Sharing of Medical Images Using Conditional GANs

Synthetic dataset for A Deep Learning Approach to Private Data Sharing of Medical Images Using Conditional GANs  Dataset specification: <ul> <li>MRI images of Vertebral Units labelled based on region</li> <li>Dataset is comprised of 10000 pairs of images and labels</li> <li>Image and label pair number k can be selected by: synthetic_dataset['images'][k] and synthetic_dataset['regions'][k]</li> <li>Images are 3D of size (9, 64, 64)</li> <li>Regions are stored as an integer. Mapping is 0: cervical, 1: thoracic, 2: lumbar</li> </ul> Arxiv paper: <a href="https://arxiv.org/abs/2106.13199">https://arxiv.org/abs/2106.13199</a> Github code: <a href="https://github.com/tcoroller/pGAN/">https://github.com/tcoroller/pGAN/</a> Abstract: Sharing data from clinical studies can facilitate innovative data-driven research and ultimately lead to better public health. However, sharing biomedical data can put sensitive personal information at risk. This is usually solved by anonymization, which is a slow and expensive process. An alternative to anonymization is sharing a synthetic dataset that bears a behaviour similar to the real data but preserves privacy. As part of the collaboration between Novartis and the Oxford Big Data Institute, we generate a synthetic dataset based on COSENTYX Ankylosing Spondylitis (AS) clinical study. We apply an Auxiliary Classifier GAN (ac-GAN) to generate synthetic magnetic resonance images (MRIs) of vertebral units (VUs). The images are conditioned on the VU location (cervical, thoracic and lumbar). In this paper, we present a method for generating a synthetic dataset and conduct an in-depth analysis on its properties of along three key metrics: image fidelity, sample diversity and dataset privacy.

opencc-by-4.0Oct 2022View details →

Amos: A large-scale abdominal multi-organ benchmark for versatile medical image segmentation (Unlabeled Data Part I)

Despite the considerable progress in automatic abdominal multi-organ segmentation from CT/MRI scans in recent years, a comprehensive evaluation of the models' capabilities is hampered by the lack of a large-scale benchmark from diverse clinical scenarios. Constraint by the high cost of collecting and labeling 3D medical data, most of the deep learning models to date are driven by datasets with a limited number of organs of interest or samples, which still limits the power of modern deep models and makes it difficult to provide a fully comprehensive and fair estimate of various methods. To mitigate the limitations, we present AMOS, a large-scale, diverse, clinical dataset for abdominal organ segmentation. AMOS provides 500 CT and 100 MRI scans collected from multi-center, multi-vendor, multi-modality, multi-phase, multi-disease patients, each with voxel-level annotations of 15 abdominal organs, providing challenging examples and test-bed for studying robust segmentation algorithms under diverse targets and scenarios. We further benchmark several state-of-the-art medical segmentation models to evaluate the status of the existing methods on this new challenging dataset. We have made our datasets, benchmark servers, and baselines publicly available, and hope to inspire future research. The paper can be found at https://arxiv.org/pdf/2206.08023.pdf In addition to providing the labeled 600 CT and MRI scans, we expect to provide 2000 CT and 1200 MRI scans without labels to support more learning tasks (semi-supervised, un-supervised, domain adaption, ...). The link can be found in: <ul> <li><a href="https://zenodo.org/deposit/7262581">labeled data (500CT+100MRI)</a></li> <li><a href="https://zenodo.org/record/7262757#.Y2iSQ9JBwYs">unlabeled data Part I (900CT)</a></li> <li><a href="https://zenodo.org/record/7295661#.Y2iR_9JBwYs">unlabeled data Part II (1100CT)</a> (Now there are 1000CT, we will replenish to 1100CT)</li> <li><a href="https://zenodo.org/record/7295816">unlabeled data Part III (1200MRI)</a></li> </ul> if you found this dataset useful for your research, please cite: <blockquote> <pre>@article{ji2022amos, title={AMOS: A Large-Scale Abdominal Multi-Organ Benchmark for Versatile Medical Image Segmentation}, author={Ji, Yuanfeng and Bai, Haotian and Yang, Jie and Ge, Chongjian and Zhu, Ye and Zhang, Ruimao and Li, Zhen and Zhang, Lingyan and Ma, Wanling and Wan, Xiang and others}, journal={arXiv preprint arXiv:2206.08023}, year={2022} }</pre> </blockquote>

opencc-by-4.0Oct 2022View details →

Amos: A large-scale abdominal multi-organ benchmark for versatile medical image segmentation (Unlabeled Data Part II)

opencc-by-4.0Aug 2019View details →

Raw Diffraction Images: Formation of a highly dense tetra rhenium cluster in a protein crystal and its implications in medical imaging.

Exploration of a “time on shelf” protein structure containing the radiopharmaceutical synthon fac-[Re(CO)3(H2O)3]+ as an in vivo reaction vessel to form tetranuclear rhenium clusters appropriate for theranostic applications.That a protein crystal can serve as a chemical reaction vessel is intrinsically fascinating. That it can produce an electron dense tetranuclear rhenium cluster compound from a rhenium tricarbonyl tribromo starting compound adds to the fascination. The cluster has been synthesised before in vitro when it formed under basic conditions. Therefore its synthesis in a protein crystal grown at pH4.5 is even more unexpected. The X-ray crystal structures presented here are for the protein hen egg white lysozyme incubated with the rhenium tricarbonyl tribromo compound for periods of 1 year and 2 years. These reveal a completed, very well resolved, tetra rhenium cluster after two years and an intermediate state after 1 year where the carbonyl ligands to the rhenium cluster are not yet clearly resolved. A dense tetra-nuclear rhenium cluster, and its technetium form, offers enhanced medical imaging contrast. The raw diffraction images for the one year and two year protein structure, described in the manuscript, is made avaliable on the Zenodo repository.

closedIPD-NOFeb 2026View details →

Imaging the Effects of Stimulant Medication on Emotional Lability in Patients With ADHD

ClinicalTrials.gov study NCT01415440. IPD Sharing: NO. Countries: 1. Publications: 1.

closedIPD-NOFeb 2026View details →

Virtual Reality Preparation for Medical Imaging

ClinicalTrials.gov study NCT03931382. IPD Sharing: NO. Countries: 1. Publications: 1.

restrictedIPD-UNDECIDEDFeb 2026View details →

PharmacofMRI (Functional Magnetic Resonance Imaging) of Anxiolytic Medications (Alprazolam)

ClinicalTrials.gov study NCT00703885. IPD Sharing: Not stated. Countries: 1. Publications: 3.