Open wanghaisheng opened 8 years ago
wanghaisheng
commented
8 years ago http://www.via.cornell.edu/databases/
| ### Chest X-ray * JSRT Digital Image Database. Digital Chest X-ray database with images containing lung nodules as well as negative cases, with ground truth location and diagnosis provided. [JSRT Database Page](http://www.jsrt.or.jp/web_data/english03.php) * SCR database: Segmentation in Chest Radiographs. Digital Chest X-ray database established to facilitate comparative studies on segmentation of the lung fields, the heart and the clavicles in standard posterior-anterior chest radiographs. [Image Sciences Institute: SCR database](http://www.isi.uu.nl/Research/Databases/SCR/) ### Computed Tomography (CT) * LIDC - NCIA Collection - Lung Image Database Consortium. Image database with lung lesions marked by up to four radiologists. [NCI-LIDC Information Page](http://imaging.cancer.gov/programsandresources/InformationSystems/LIDC). Size information on this dataset is provided by the VIA group here: [LIDC Nodule Size Report](http://www.via.cornell.edu/lidc/). The images and main documentation are available at [The Cancer Imaging Archive (TCIA)](http://cancerimagingarchive.net/). * RIDER - NCIA Collection - Reference Image Database to Evaluate Response. Image archive of CT lung cancer patients followed during treatment. [RIDER - White Paper](http://imaging.cancer.gov/images/documents/RIDER-executive-summaryA_071306.pdf) The images and main documentation are available at [The Cancer Imaging Archive (TCIA)](http://cancerimagingarchive.net/). * [ELCAP public database of whole lung CT images](./lungdb.html) An image database of whole-lung low-dose CT images of the chest with the locations of all nodules marked. * [Public Database to Address Drug Response](./crpf.html) An image database with small datasets of well documented CT images of the chest ### Magnetic Resonance Images * BrainWeb: Simulated Brain Database. Contains simulated, 3D MR data using normal and multiple sclerosis models with different acquisition parameters. [BrainWeb: SBD](http://www.bic.mni.mcgill.ca/brainweb/) ### Mammography * DDSM: Digital Database for Screening Mammography. Contains a large number of cases with both normal and abnormal findings (and associated ground truth). [USF DDSM Homepage](http://marathon.csee.usf.edu/Mammography/Database.html) * Mini-MIAS (Mammographic Image Analysis Society). Contains cases with and location information of the abnormality. [Mini-MIAS](http://peipa.essex.ac.uk/info/mias.html) ### Retinopathy * DIARETDB1 - Standard Diabetic Retinopathy Database. Database for benchmarking diabetic retinopathy detection from digital images. Offers a standardized testing protocol. [Website](http://www.it.lut.fi/project/imageret/diaretdb1/) * DRIVE: Digital Retinal Images for Vessel Extraction. Database established to facilitate comparative studies on segmentation of blood vessels in retinal images. [Website](http://www.isi.uu.nl/Research/Databases/DRIVE/) ### Virtual Colonoscopy * NCIA Collection: Virtual Colonoscopy - Database for colonic polyp detection from CT with MS Access relational database file to aid in case selection found at [NCIA Collections Page](https://wiki.nci.nih.gov/x/lRWy). _*This collection is made available from the Walter Reed Army Medical Center Virtual Colonoscopy Collection in collaboration with National Cancer Institute, NIH: please note the citation requirements on the NCIA collections page._ ### CT Colonography * TCIA Collection: ACRIN CT Colonography Collection for colonic polyp detection from CT with XLS sheets that provide polyp descriptions and their location within the colon segments ( [TCIA Collections Page](https://wiki.cancerimagingarchive.net/display/Public/Collections)). ### Multimodality * Standardized Evaluation Methodology for 2D-3D Registration Database. Contains multi-modality datasets, centers of rotation, evaluation criteria, and starting positions for the evaluation of registration methods. [Image Sciences Institute: Registration database](http://www.isi.uu.nl/Research/Databases/GS/) * TCIA Collection: PET/CT phantom scan collection. A resource for increased quantitative understanding of machine acquisition, analytic reproducibility and image processing. [TCIA Collections Page](https://wiki.cancerimagingarchive.net/display/Public/Collections) (in transfer from NBIA) |
| ### Computed Tomography (CT) * EXACT09\. Challenge for the automated extraction of airways from CT data. [EXACT09 Homepage](http://image.diku.dk/exact/index.php) * ANODE09 Study. Challenge for the CAD detection of lung lesions from whole-lung CT data. [ANODE09 Homepage](http://anode09.isi.uu.nl/index.php) * BIOCHANGE 2008 PILOT. Challenge for the evaluation of change measurement algorithm. [BIOCHANGE 2008 Pilot Homepage](http://www.itl.nist.gov/iad/894.05/biochange2008/Biochange2008-webpage.htm) _*Note: Uses publicly accessible NCIA-RIDER and FDA phantom data._ * [VOLCANO'09 Nodule Change Challenge](/challenge/) the evaluation of size change in pulmonary nodules. * Coronary Artery Algorithm Evaluation Framework. Challenge based on the extraction of coronary artery centerlines from CTA data. [Home Page](http://coronary.bigr.nl) * Liver Tumor Segmentation 08\. Challenge based on the segmentation of liver lesions from contrast enhanced CT. [Challenge Homepage](http://lts08.bigr.nl) ### Magnetic Resonance Images * MS lesion segmentation challenge 08\. Challenge for the segmentation of brain lesions from MR imagery. [MSseg08 Homepage](http://www.ia.unc.edu/MSseg/) * CAUSE07 - Caudate Segmentation Evaluation 2007\. Challenge for the segmentation of the caudate nucleus from brain MRI scans. [CAUSE07 Homepage](http://www.cause07.org) ### Retinopathy * ROC-Retinopathy Online Challenge. Challenge based on detecting microaneurysms and dot hemorrhages for diabetic retinopathy screening. [The ROC Website](http://roc.healthcare.uiowa.edu/) ### Grand Challenges List * A list of grand challenges for medical image analysis is maintained at [grand-challenge.org](http://grand-challenge.org) |
| [ECLAP public database of whole lung CT images](./lungdb.html) | 50 cases of low-dose thin-slice chest CT images with annotations for small nodules |
| [Public Database to Address Drug Response](./crpf.html) | Over 100 cases of CT chest images illustrating the spectrum of nodule presentations together with a range of computer analysis methods. |
| [VOLCANO'09 Nodule Change Challenge](/challenge/) | A set of benchnmark pairs CT images of pulmonary nodules to provide a challenge for the evaluation of nodule change in size measurement methods |
| [LIDC database size index list](/lidc/) | Standardized nodule lists and spreadsheets for the LIDC public image database |
| The Lung Image Database Consortium [(LIDC)](http://imaging.cancer.gov/programsandresources/InformationSystems/LIDC) |
| The Image Database Resource Initiative [(IDRI)](http://imaging.cancer.gov/programsandresources/InformationSystems/) |
| The Reference Image Database to Evaluate Response [(RIDER)](http://imaging.cancer.gov/images/documents/RIDER-executive-summaryA_071306.pdf) |
| The public databases for these projects can be accessed through the [The Cancer Imaging Archive (TCIA)](http://cancerimagingarchive.net/). |
wanghaisheng
commented
8 years ago wanghaisheng
commented
8 years ago 2008 MICCAI MS Lesion Segmentation Challenge (National Institutes of Health Blueprint for Neuroscience Research)
Annotated Spine CT Database for Benchmarking of Vertebrae Localization, 125 patients, 242 scans (Ben Glockern)
Cavy Action Dataset - 16 sequences with 640 x 480 resolutions recorded at 7.5 frames per second (fps) with approximately 31621506 frames in total (272 GB) of interacting cavies (guinea pig) (Al-Raziqi and Denzler)
Computed Tomography Emphysema Database (Lauge Sorensen)
CRCHistoPhenotypes - Labeled Cell Nuclei Data - colorectal cancer histology images consisting of nearly 30,000 dotted nuclei with over 22,000 labeled with the cell type (Rajpoot + Sirinukunwattana)
Dermoscopy images (Eric Ehrsam)
DIADEM: Digital Reconstruction of Axonal and Dendritic Morphology Competition (Allen Institute for Brain Science et al)
DIARETDB1 - Standard Diabetic Retinopathy Database (Lappeenranta Univ of Technology)
DRIVE: Digital Retinal Images for Vessel Extraction (Univ of Utrecht)
Leaf Segmentation ChallengeTobacco and arabidopsis plant images (Hanno Scharr, Massimo Minervini, Andreas Fischbach, Sotirios A. Tsaftaris)
MiniMammographic Database (Mammographic Image Analysis Society)
MIT CBCL Automated Mouse Behavior Recognition datasets (Nicholas Edelman)
Moth fine-grained recognition - 675 similar classes, 5344 images (Erik Rodner et al)
Mouse Embryo Tracking Database - cell division event detection (Marcelo Cicconet, Kris Gunsalus)
OASIS - Open Access Series of Imaging Studies - 500+ MRI data sets of the brain (Washington University, Harvard University, Biomedical Informatics Research Network)
Plant Phenotyping Datasets - plant data suitable for plant and leaf detection, segmentation, tracking, and species recognition (M. Minervini, A. Fischbach, H. Scharr, S. A. Tsaftaris)
Retinal fundus images - Ground truth of vascular bifurcations and crossovers (Univ of Groningen)
Spine and Cardiac data (Digital Imaging Group of London Ontario, Shuo Li)
Univ of Central Florida - DDSM: Digital Database for Screening Mammography (Univ of Central Florida)
VascuSynth - 120 3D vascular tree like structures with ground truth (Mengliu Zhao, Ghassan Hamarneh)
York Cardiac MRI dataset (Alexander Andreopoulos)The USC-SIPI Image Database The USC-SIPI image database is a collection of digitized images. It is maintained primarily to support research in image processing, image analysis, and machine vision. The first edition of the USC-SIPI image database was distributed in 1977 and many new images have been added since then.
The database is divided into volumes based on the basic character of the pictures. Images in each volume are of various sizes such as 256x256 pixels, 512x512 pixels, or 1024x1024 pixels. All images are 8 bits/pixel for black and white images, 24 bits/pixel for color images. The following volumes are currently available:
Textures Brodatz textures, texture mosaics, etc.
Aerials High altitude aerial images
Miscellaneous Lena, the mandrill, and other favorites
Sequences Moving head, fly-overs, moving vehiclesFile Format and Names
Note: It is the database user's responsibility to figure out how to read the images into whatever computer they will be using, and how to access the files from within application programs. USC-SIPI does not have the resources to provide assistance in these areas. If you are in doubt about whether or not you will be able to read the images on your computer, please check with your system managers and show them the description of the database and the image formats.
All images in the database are currently stored in TIFF format. Some information about the TIFF format is available from http://www.awaresystems.be/imaging/tiff.html and from Adobe Systems. The "libtiff" library of C functions for reading and writing TIFF images is available from http://www.remotesensing.org/libtiff/. The "netpbm" collection of image format conversion programs (http://netpbm.sourceforge.net/) can convert between TIFF and many different formats.
Previous versions of the database that were only distributed on magnetic tape used a raw binary format for the data instead of the TIFF format. The TIFF files in the current edition can be converted back to a raw binary format using the TIFF software available at the above sites.
A sample C program for converting an image from TIFF to raw format on a Unix system is available (tiff2raw.c). This program should convert the color (24 bits/pixel) and grayscale (8 bits/pixel) images in the database into the raw binary format files. This program requires the "libtiff" library available at the site mentioned above. Note that while we believe this program works properly on the TIFF images in the database, it has NOT been tested on other TIFF images and may not convert them correctly.
Many of the images in the database have numerical filenames such as 4.2.04. These relate to an numbering scheme that was used with an earlier edition of the database that was released in 1981. File Checksums
A list of checksum values for all the files in the database is available. This list can be used to check for any corruption of the image data after downloading the files. Copyright Information
The images in the USC Image Database are intended for research purposes. USC-SIPI does not own the copyright of most of the images and the copyright status of many of the images is unknown. For more information on the copyright status of the images, please click here. If you plan on using any of the images in a publication, please read the copyright information before committing to using any of the images. Database Catalog
A printed description of the USC-SIPI Image Database, including reduced resolution pictures of most of the images in the database, is available in PDF format for downloading and viewing using Adobe Acrobat. The file size is about 4mb. A printed copy of the catalog can also be obtained by contacting USC-SIPI. Other Means of Distribution
Copies of the USC-SIPI Image Database can also be purchased on CD-R. Contacting USC-SIPI
Technical questions about the database images, the image format, or problems obtaining the images should be directed to the database editor, Allan Weber, , 213-740-4147.
Inquiries about purchasing the database or other USC-SIPI services can be sent to , 213-740-4145.
wanghaisheng
commented
8 years ago Contents:
IntroductionImages are an essential part of medical education and practice. Traditionally, drawings and illustrations have been an integral tool for sharing medical knowledge. Examples date back to Leonardo DiVinci's 16th century drawings of cadaveric dissections. Over the centuries, new technologies have lead to an increased number of formats available for capturing and displaying medical images. Image formats have evolved from drawings and engravings, to analog photographs and videotape, to digital image files.
Before an image collection can serve as an educational tool, the student, researcher or health care practitioner must have access to it. Slide and photograph collections, which have been heavily used in fields such as radiology and pathology, are laden with access and management problems. These problems include both physical and intellectual access, the high cost of duplication and the inability to easily integrate and disseminate new materials.
The combination of digital image formats and networked computer environments provides the potential to overcome the traditional problems of medical image collections. This paper reports on medical image databases on the World Wide Web (WWW). An annotated list of Web sites containing medical images is also provided.
Problem-Solving with Digitization and the WWWThis section addresses each of these problems and how they may be solved by using a digitized image collection available through the WWW.
Physical and intellectual access
Both physical and intellectual access present problems to the users of image collections. Photograph and slide collections are typically stored in a single area within the university. Storage may be at the medical department, or a more centralized location such as the medical library or learning resource center. In each of the cases the user and the collection must be physically brought together. This requirement severely restricts the user who may wish to study at home or at a location miles away from the collection. Additional restrictions are placed on the user if the image collection can only be accessed during limited operating hours or if the collection is in use by another person.
Once the user and the collection are brought together, intellectual access, that is retrieving the set of images relevant to the user's needs, presents another barrier. Browsing through large sets of photographs or trays of slides can be extremely time consuming. Even when cataloging is provided, it is often difficult to determine the usefulness of an image from a textual description.
Providing an image database on-line alleviates the problems of physical access. On-line image files can be viewed from computer terminals directly connected to the university network from anywhere on campus, from a remote connection at home or from anywhere else in the world. Connecting to the WWW with browsers such as Mosaic and Netscape allows for seamless downloading of image files. Furthermore, files can be available 24 hours a day, seven days a week and available simultaneously to many users.
Within the Web environment, intellectual access problems can be overcome by using combinations of search engines and thumbnail size images. For example, users could fill in a key word or string and retrieve a set of small images to browse. The set of small images can, in turn, be clicked to retrieve the document containing that image and its associated text. This strategy saves time by allowing the user to preview small, low resolution images prior to downloading large, high quality image files.
Deterioration of the collection and duplication costs
The quality of photographs and slides degrades over time and collections need to be replaced. Depending on the type of emulsion used, films turn brown, pink and may fade away altogether. Photographs and slides can also be easily damaged; a fingerprint or a drop of water can permanently ruin the image. Additionally, photos and slides can be lost and stolen.
Replacement and duplication of film collections is expensive. The cost of copying a 1000-case teaching file can be upwards of $10,000 for the film alone. A digital image can be duplicated swiftly and economically with a few simple keystrokes. The digital copy is an exact duplicate of the original with no loss of quality regardless of how many copies are made.
Inability to distribute collection updates
A well maintained teaching file should be dynamic with new cases added at regular intervals. Photograph and slide collections, however, become static entities as soon as they are distributed. Digital image files stored on a server have the power to be updated with ease. Changes that are made to a file are immediately available to the networked users.
In addition to overcoming the above problems, digital images on the WWW provide many distinct advantages over film-based collections. Digital images can easily be manipulated and enhanced. They can be saved at various sizes and resolutions, annotated and filtered. Global access to the teaching file can be bi-directional; contributions to the file can be transmitted over the network and incorporated into the file.
The Downside of On-line Image FilesProblems related to Navigation
A common complaint regarding the WWW is the lack of indexing and the inability to find useful information. This criticism applies to medical image files on a macro and micro level. On the broad, macro scale, wading through long lists and directories of potential sites can be time consuming. Search engines and directories may expedite the process but Web sites vary greatly in terms of content, quality and size, therefore users may need to explore several sites before finding an appropriate database of images.
On a micro level, navigating through a Web site may leave the user frustrated. Links within a Web site can be organized in a hierarchical arrangement, a flat arrangement or any permutation of the two. Indexes and tables of contents may not be logically arranged or may simply not be included. Terms chosen as links may be misleading or ambiguous and the user may follow several wrong paths before arriving at the desired destination.
Document boundaries present another navigational problem. Web sites can be comprised of any number of files. Well designed sites include consistent formatting and layout, graphics and metadata to maintain the look and feel of the parent document. When these cues are missing, it can be difficult to determine if there is a connection between documents or if a link has been followed outside of the initial Web site.
Another problem related to formatting is the over-use or inappropriate use of HTML tags. Poor design or ill attempts to enhance a document can obliterate the content and make the document difficult to use.
Problems related to Images
A further issue affecting the ease-of-use of medical image Web sites is related to the large size of image files. Good quality images are large in terms of transmission time across a network. In designing Web sites, it necessary to consider the impact of how the images are incorporated into the document. Many sites do not include thumbnail versions of the image, thus forcing the user to wait for lengthy downloads before its value can be determined.
In any image database, the images themselves do not provide descriptive information. This becomes salient on the WWW because images can easily become detached from the textual portion of the document, as when images are copied and saved to a remote terminal. An advantage of digital images is the ability to attach annotations and watermarks onto an image. Watermarks create a digital signature which help to protect the owner of the image from copyright violations. At this time, attaching descriptive text and watermarks to medical images on the WWW is not standard practice.
Image File Formats & Medical Image ModalitiesImage file formats available for storing digital images include TIFF, GIF, JFIF, SPIFF, PICT, TGA, EPS, CGM, and Photoshop. GIF, TIFF, and JPEG formats are compatible with the World Wide Web. For an explanation of file formats, view Working with Images by Lee Liming or Graphics maintained by Martin Reddy.
Sophisticated image production has improved diagnostic procedures in many medical fields, such as radiology and cardiology. For example, physicians can view images of the brain with CT (computed tomography) and MRI (magnetic resonance imaging), and can observe organ physiology such as cardiac circulation with angiography.
The Web sites listed in the following section, Medical Image Sites on the Web, contain various types of medical images such as X-ray, PET and MRI. Brief descriptions are presented here to introduce the basic concept underlying each imaging modality. The descriptions are presented in order of increasing technical complexity.
Medical Image Sites on the WebMany fields of medicine have produced image databases for the internet. Fields that have traditionally been heavily image oriented, such as radiology and pathology are highly represented on the World Wide Web. Other medical disciplines represented on the WWW include dermatology, ophthalmology, pediatrics and more.
This list of medical image Web sites illustrates a cross-section of what's out there and is not intended to be exhaustive. A well maintained, extensive list of internet medical resources produced by Gary Malet and Lee Hancock can be viewed at Medical Matrix. Pointers to more medical image sites can be found at Medical Images and Multimedia.
Go to:
Anatomy | Cardiology | Dermatology | Hematology | Nuclear Medicine | Neurology | Obstetrics & Gynecology | Oncology | Ophthalmology | Pathology | Pediatrics | Pulmonology | Radiology |
End of List
Go to:
Anatomy | Cardiology | Dermatology | Hematology | Nuclear Medicine | Neurology | Obstetrics & Gynecology | Oncology | Ophthalmology | Pathology | Pediatrics | Pulmonology | Radiology |
End of List
Go to:
Anatomy | Cardiology | Dermatology | Hematology | Nuclear Medicine | Neurology | Obstetrics & Gynecology | Oncology | Ophthalmology | Pathology | Pediatrics | Pulmonology | Radiology |
End of List
ConclusionDisseminating medical images across a networked environment has great potential for sharing knowledge among medical professionals. Improvements in technology will undoubtedly alleviate the sluggish transmission of large images as well as reduce the cost of creating and storing digital image files. Additional improvements will result form the adoption of standards which facilitate the user in finding appropriate information and navigating through Web sites.
References Go to: top of page
Produced for ILS 603, taught by Howard Besser.
Copyright 1995, nscherer
wanghaisheng
commented
8 years ago wanghaisheng
commented
8 years ago Open-Access Medical Image Repositories
|
Sites that list and/or host multiple collections of data:
* [National Biomedical Imaging Archive (NBIA):](https://imaging.nci.nih.gov/ncia/)
* Lung Image Database Consortium (LIDC)
* Reference Image Database to Evaluate Response (RIDER)
* Breast MRI
* Lung PET/CT
* Neuro MRI
* CT Colongraphy
* Virtual Colonoscopy
* Osteoarthritis Initiative
* PET/CT phantom scan collection
* [BIRN / XNat](http://www.birncommunity.org/resources/data/)
* High SNR Healthy Volunteer DTI Calibration Dataset
* Morphometry BIRN Multi-site Multi-session Structural MRI Data
* Open Access Structural Imaging Series (OASIS)
* Duke Center for In-Vivo Microscopy High Resolution MRI Images
* [MIDAS](http://www.insight-journal.org/midas/)
* [National Alliance for Medical Image Computing (NAMIC)](http://www.insight-journal.org/midas/community/view/17)
* Lupus white matter lesions
* Brain MRI: 2-4 years old
* Prostate
* [NLM: Imaging Methods Assessment and Reporting](http://www.insight-journal.org/midas/community/view/15)
* Liver tumors with segmentations
* [100 Healthy Brain MRI: 18-90 years old](http://www.insight-journal.org/midas/community/view/21)
* [UCI Machine Learning Repository:](http://archive.ics.uci.edu/ml/) The father of internet data achives for all forms of machine learning.
* [Cornell Visualization and Image Analysis (VIA) group:](http://www.via.cornell.edu/databases/) Provides a list of available databases, many of which are also listed here.
* [UT Health Scient Center Image Collections:](http://www.library.uthscsa.edu/find/databases.cfm?Category=Image%20Collections) List of medical images, atlases, and databases available on the web.
* [OmniMedicalSearch.com: Medical Image Databases & Libraries](http://www.omnimedicalsearch.com/image_databases.html)
Sites dedicated to specific data collections:
* [Digital Database for Screening Mammography (DDSM):](http://marathon.csee.usf.edu/Mammography/Database.html) Large collection with normal and abnormal findings and ground truth.
* [Digital Retinal Images for Vessel Extraction (DRIVE):](http://www.isi.uu.nl/Research/Databases/DRIVE/) Digital images and expert segmentations of retinal vessels.
* [Japanese Society of Radiological Technology (JSRT) Database:](http://www.jsrt.or.jp/web_data/english03.php) Digital Chest X-ray images with lung nodule locations, ground truth, and controls.
* [Segmentation in Chest Radiographs (SCR) database:](http://www.isi.uu.nl/Research/Databases/SCR/) Digital Chest X-ray images with segmentations of lung fields, heart, and clavicles.
* [Public Lung Database to Address Drug Response:](http://www.via.cornell.edu/crpf.html) Well documented chest CT images.
* [Mammographic Image Analysis Society (mini-MIAS) Database:](http://peipa.essex.ac.uk/info/mias.html) Mammographic images and markup.
* [Standard Diabetic Retinopathy Database (DIARETDB1):](http://www2.it.lut.fi/project/imageret/diaretdb1/) Digital retinal images for detecting and quantifying diabetic retinopathy.
* [SpineWeb](http://spineweb.digitalimaginggroup.ca/): SpineWeb is an online collaborative platform for everyone interested in research on spinal imaging and image analysis.
Data for registration method development and evaluation:
* [Retrospective Image Registration Experiment (RIRE)](http://www.insight-journal.org/rire/)
* [Standardized Evaluation Methodology for 2D-3D Registration](http://www.isi.uu.nl/Research/Databases/GS/)
Simulated and phantom data available for public use:
* [BrainWeb:](http://mouldy.bic.mni.mcgill.ca/brainweb/) Simulated brain MR database.
Data from CAD competitions, some is available for public use:
* Portal for over 100 grand challenges in medical imaging:
* [www.grand-challenge.org](http://www.grand-challenge.org)
* CAUSE07: Segment the caudate nucleus from brain MRI.
* BIOCHANGE 2008 PILOT: Measure changes.
* MS lesion segmentation challenge 08 Segment brain lesions from MRI.
* Liver Tumor Segmentation 08 Segment liver lesions from contrast enhanced CT.
* EXACT09: Extract airways from CT data.
* ANODE09: Detect lung lesions from CT.
* VOLCANO09: Quantify changes in pulmonary nodules.
* Coronary Artery Algorithm Evaluation Framework: Extract coronary artery centerlines from CTA data.
* ROC-Retinopathy Online Challenge: Detect microaneurysms for diabetic retinopathy screening.
|
wanghaisheng
commented
8 years ago Finding, Downloading, and Managing Medical Images
wanghaisheng
commented
8 years ago SCR database: Segmentation in Chest Radiographs
The automatic segmentation of anatomical structures in chest radiographs is of great importance for computer-aided diagnosis in these images. The SCR database has been established to facilitate comparative studies on segmentation of the lung fields, the heart and the clavicles in standard posterior-anterior chest radiographs.
In the spirit of cooperative scientific progress, we freely share the SCR database and are committed to maintaining a public repository of results of various algorithms on these segmentation tasks. On thes pages, instructions can be found on downloading the database and uploading results, and benchmark results of various methods can be inspected.
The data included in this database can be used, free of charge, for research and educational purposes. Redistribution and commercial use is prohibited. Any researcher reporting results which use this database must acknowledge the SCR database. We kindly request you to do so by citing this publication:
In addition, you should also cite the following publication which describes the JSRT database (images were taken from this database):
We appreciate to hear about any publications that use the SCR database. Feedback on the database and this website is also welcome. The person to contact is Bram van Ginneken.
All chest radiographs are taken from the JSRT database. This is a publicly available database with 247 PA chest radiographs. In each image the lung fields, heart and clavicles have been manually segmented to provide a reference standard.
For more information, see the data description page.
The download page gives instructions on downloading the SCR database.
The result browser allows interactive viewing of images and segmentations and is also a useful way to explore the database.
A tabular overview of the results of various algorithms on each objects can be found here.
The data has been subdivided in two folds. Fold 1 consists of all odd numbered images in the JSRT database. Fold 2 consists of all even numbered images. To ensure the integrity of results, you should use the images and reference standard in one fold to train and tune your algorithms, and apply the algorithm to segment the images in the other fold.
We invite other researchers to run their segmentation algorithms against the SCR database. If the results are sent to us, we will then run our evaluation software, and report the algorithm's performance on this site. Currently our evaluation is performed on hard (binary) classification images, which should be supplied in the same format as the binary masks that can be downloaded. Additionally, a literature pointer, article, report or web site that we can refer to for a description of the applied algorithm is requested.
To upload your results, contact Bram van Ginneken by e-mail.
wanghaisheng
commented
8 years ago wanghaisheng
commented
8 years ago Databases you can use for benchmarking
El Salvador Atlas of Gastrointestinal VideoEndoscopy Images and Videos of his-res of studies taken from Gastrointestinal Video endoscopy. (Format: jpg, mpg, gif)
The Mammographic Image Analysis Society (MIAS) mini-database.
Mammography Image Databases 100 or more images of mammograms with ground truth. Additional images available by request, and links to several other mammography databases are provided. (Format: homebrew)
NLM HyperDoc Visible Human Project color, CAT and MRI image samples - over 30 images (Format: jpeg)
University of Massachusetts Vision Image Archive Large image database with aerial, space, stereo, medical images and more. (Format: homebrew) Radiology Image Database The National Library of Medicine presents MedPix®
The National Library of Medicine is pleased to announce the launch of MedPix®, a free online medical image database. The URL is https://medpix.nlm.nih.gov/.
The MedPix collection categorizes and classifies the image and patient data for each of several subsets of image database applications (e.g. radiology, pathology, ophthalmology, etc.). The content material is both high-quality and high-yield and includes both common and rare conditions. Most cases have a proven diagnosis (pathology, clinical follow-up). The teaching file cases are peer-reviewed by an Editorial Panel.
As a public education service, the NLM and MedPix provide the storage service, indexing, and Web server hosting. Individuals as well as institutions may participate. Contributed content may be copyrighted by the original author/contributor. No additional software required.
wanghaisheng
commented
8 years ago Can anyone suggest medical image database links?
http://www.humanconnectome.org
http://www.medinfo.cs.ucy.ac.cy/index.php/downloads/datasets
Still under construction but you have some prostate MRI dataset:
DDSM: Digital Database for Screening Mammography
http://marathon.csee.usf.edu/Mammography/Database.html
Breast Cancer Wisconsin (Diagnostic) Data Set
http://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+%28Diagnostic%29
Hope this helps: DICOM sample image datasets
http://www.osirix-viewer.com/datasets/Belarus dataset have both chest X-rays and CT scans of the same patient. you can download from their website:
http://www.ctisus.com/ LIDC-IDRI https://wiki.cancerimagingarchive.net/display/Public/LIDC-IDRI
wanghaisheng
commented
8 years ago http://www.learningradiology.com/images/chestimages1/chestgallerypages/index.html http://www.radiologyassistant.nl/en/p497b2a265d96d/chest-x-ray-basic-interpretation.html http://www.glowm.com/atlas_page/atlasid/chestXray.html http://www.chestx-ray.com/index.php/education/normal-cxr-module-train-your-eye#!16
wanghaisheng
commented
8 years ago http://www.cardiacatlas.org/studies/ AMRG Cardiac Atlas The AMRG Cardiac MRI Atlas is a complete labelled MRI image set of a normal patient's heart acquired with the Auckland MRI Research Group 's Siemens Avanto scanner. The atlas aims to provide university and school students, MR technologists, clinicians... Congenital Heart Disease (CHD) Atlas Congenital Heart Disease (CHD) Atlas The Congenital Heart Disease (CHD) Atlas represents MRI data sets, physiologic clinical data and computer models from adults and children with various congenital heart defects. The data have been acquired from several clinical centers including Rady... DETERMINE DETERMINE DETERMINE (Defibrillators to Reduce Risk by Magnetic Resonance Imaging Evaluation) is a prospective, multicenter, randomized clinical trials in patients with coronary artery diseases and mild-to-moderate left ventricular dysfunction. The primary objective... MESA MESA MESA (Multi-Ethnic Study of Atherosclerosis ) is a large-scale cardiovascular population study (>6,500 participants) conducted in six centres in the USA. It aims to investigate the manifestation of subclinical to clinical cardiovascular disease before... SCMR Consensus Data SCMR Consensus Data The SCMR Consensus Dataset is a set of 15 cardiac MRI studies of mixed pathologies (5 healthy, 6 myocardial infarction, 2 heart failure and 2 hypertrophy), which were acquired from different MR machines (4 GE, 5 Siemens, 6 Philips). The main objectives... Sunnybrook Cardiac Data Sunnybrook Cardiac Data The Sunnybrook Cardiac Data (SCD) , also known as the 2009 Cardiac MR Left Ventricle Segmentation Challenge data, consist of 45 cine-MRI images from a mixed of patients and pathologies: healthy , hypertrophy , heart failure with infarction and heart...
wanghaisheng
commented
7 years ago openI X 胶片和报告 脱敏最佳实践 Preparing a collection of radiology examinations for distribution and retrieval.
Q: What is Open-i?
A: Open-i service of the National Library of Medicine enables search and retrieval of abstracts and images (including charts, graphs, clinical images, etc.) from the open source literature, and biomedical image collections. Searching may be done using text queries as well as query images. Open-i provides access to over 3.7 million images from about 1.2 million PubMed Central® articles; 7,470 chest x-rays with 3,955 radiology reports; 67,517 images from NLM History of Medicine collection; and 2,064 orthopedic illustrations.
Longer descriptions are available in a research article "Design and Development of a Multimodal Biomedical Information Retrieval System" ( http://lhncbc.nlm.nih.gov/system/files/pub2012019.pdf ) and a technical report ( http://lhncbc.nlm.nih.gov/system/files/tr2010002.pdf )
Q: Are the images in Open-i collection in public domain? May I use a particular image for my project?
A: Open-i images are from one of the following sources:
The Open Access Subset of PubMed Central (PMC), a free full-text archive of biomedical and life sciences journal literature at the U.S. National Library of Medicine.
The Indiana University hospital network.
The Orthopedic Surgical Anatomy Teaching Collection (http://digitallibrary.usc.edu/cdm/landingpage/collection/p15799coll50) at the USC Digital Library (http://digitallibrary.usc.edu/).
Images from the History of Medicine Division (https://www.nlm.nih.gov/hmd/) from the U.S. National Library of Medicine.
MedPix (https://medpix.nlm.nih.gov/)
Reuse of Open-i images is determined by the license type of the image. A link to the applicable license type, if available, may be found below the individual Open-i image on the detailed view page.
Please refer to the following web page for information on the PMC open access subset and the PMC Copyright Notice:
http://www.ncbi.nlm.nih.gov/pmc/tools/openftlist/
http://www.ncbi.nlm.nih.gov/pmc/about/copyright/
If copyright restrictions for an image in the PubMed Central collection are not specified or are unclear, please contact directly the journal directly that published the image.
To navigate to the full version of the paper at the journal publisher site, please follow the link to PubMed® located below the title. When viewing the abstract in PubMed, use the links to the full text in the upper right corner to get to the journal that published the paper containing the image. If a link to HTML is available below the title, it will bring you to the journal directly.
Images from the Indiana University hospital network are distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ).
Images from the Orthopaedic Surgical Anatomy Teaching Collection at the USC Digital Library have the following Rights statement:
Distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 2.5 License (http://creativecommons.org/licenses/by-nc-sa/2.5/) that permits unrestricted use, distribution and reproduction in any medium, provided the work is attributed to Irving Rehman, Ph.D., F.I.C.S., and Chadwick F. Smith, M.D., Ph.D., F.A.C.S., F.I.C.S., and source in the manner specified by the publisher.
Please refer to the following web page for copyright information for Images from the History of Medicine:
https://www.nlm.nih.gov/hmd/copyright/
MedPix® images and case materials were contributed by many individuals. They are organized, reviewed, approved, and curated free of charge for your personal use and for local teaching at your institution - including distribution of handouts and syllabi. For anything other than personal use, you should respect the original contributor and contact them for additional permission requests.
Q: Where can I get the Chest X-ray images in Open-i ?
A: The chest x-ray images from the Indiana University hospital network are available here:
PNG images: Link
DICOM images: Link
We request not to share the datasets outside of your research group/organization, but forward interested researchers/ new requests to us. Also, please inform us if you find errors or inconsistencies in the data. Finally, we ask that publications resulting from the use of this data attribute the data to National Library of Medicine, National Institutes of Health, Bethesda, MD, USA, and cite this article: https://lhncbc.nlm.nih.gov/publication/pub9189
Q: I have heard about the Tuberculosis collection. Where can I get those images ?
A: The Communications Engineering Branch at the Lister Hill National Center for Biomedical Communications, an intramural division of the National Library of Medicine has the following de-identified image data sets of chest x-rays (CXRs) available to the research community. Both sets contain normal as well as abnormal x-rays, with the latter containing manifestations of tuberculosis. We request not to share these datasets outside of your research group/organization, but forward interested researchers / new requests to us. Also, please inform us if you find errors or inconsistencies in the data. Finally, we ask that publications resulting from the use of this data attribute the data to National Library of Medicine, National Institutes of Health, Bethesda, MD, USA, and cite this article: https://lhncbc.nlm.nih.gov/publication/pub9356
Montgomery County X-ray Set: X-ray images in this data set (Download here: Link) have been acquired from the tuberculosis control program of the Department of Health and Human Services of Montgomery County, MD, USA. This set contains 138 posterior-anterior x-rays, of which 80 x-rays are normal and 58 x-rays are abnormal with manifestations of tuberculosis. All images are deidentified and available in DICOM format. The set covers a wide range of abnormalities, including effusions and miliary patterns. The data set includes radiology readings available as text file.
Shenzhen Hospital X-ray Set / China data set: X-ray images in this data set (Download here: Link) have been collected by Shenzhen No.3 Hospital in Shenzhen, Guangdong providence, China. The x-rays were acquired as part of the routine care at Shenzhen Hospital. The set contains images in JPEG format. There are 340 normal x-rays and 275 abnormal x-rays showing various manifestations of tuberculosis.
Q: How do I use an image as query?
A: There are two ways to search using images:
Drag and drop the image file onto the Open-i search or results page to initiate an image search. Supported file types are .jpeg, .jpg, .gif and .png.
Upload an image file using the “Query by Image” link in the upper right corner.
Q: What is the difference between the Citation List view and Image Grid?
A: The citation list view shows the best match image on the left along with citation information on the right. The grid view displays all matching images from an article. You may switch between views by clicking on the appropriate icon next to “View as” area to the right of the search box.
Q: How do I rank my results by date?
A: Open the Rank By section in the “Limits” bar. Click on Newest or Oldest. This only gives more weight to the date; it does not sort by date.
Q: How do I use limits?
A: By default, the “exclude graphics” limit is applied to the search, which excludes charts, graphs, drawings, etc. All subsets and specialties are searched. Text is searched in all fields. Results are ranked in relevance order. To filter the results, you may select one or more categories in the “Limits” bar. To view or remove selected filters, use the “Selected Limits” area above the “limits” bar.
Q: How do I open “View Article” results in a new window?
A: Control-Click on the link to open and view the article in a new browser window.
Q: How do I open “View Similar Images In” results in a new window?
A: Control-Click on the link to open and view the results in a new browser window.
Q: When my browser window is narrow I lose the View Similar Images and View Article links in list view?
A: When the window is narrow, Open-i responsive design puts the user in mobile mode. Click, tap, or swipe the three vertical dots to view those links.
Q: In Chrome, my video doesn't play somtimes, what do I do?
A: We have a known issue for some videos in Chrome. Please do a Ctrl-Shift-R to refresh the page and the video will play.
Q: Can I do an exact search? I am looking for AID and don't want to see papers about HIV/AIDS.
A: By default, the search terms are expanded with UMLS synonymy and the singular/plural forms, so aid is expanded to HIV/AIDS.
To avoid expansion, use NO_EXPAND (all upper case) + search term, e.g., NO_EXPAND AID If you want singular/plural expansion, but not UMLS synonymy, use TERM_EXPAND + search terms. For example, NO_EXPAND heart attack finds 89 articles; TERM_EXPAND heart attack finds 118, and searching for heart attack with default setting results in 2805 articles.
NIH提供的癌症测试数据集