Image is Everything: Finding and Using Medical Images on the Web

March 4, 2009 at 9:37 am 2 comments

Images (graphics, illustrations, photographs, etc.) are visual representations; they help convey ideas and concepts.  Research has shown that humans are better at retaining information communicated in images than in text.[1]  The formats in which images have been created have evolved over the centuries, from static images on cave walls, papyrus, paper, photographic and radiographic film to two- and three-dimensional dynamic computerized images.[2]  The mechanisms for delivering, storing and retrieving the images have evolved as well.  From our digital cameras and cell phones, we can download images to web servers, PDAs, MP3 players, or attach them to e-mail messages.  We can go to the Internet to search, access and download digitized or “born-digital” images.  We no longer have to rely on printed books, magazines nor travel to far away places to view and appreciate prehistoric cave paintings (http://www.culture.gouv.fr/culture/arcnat/lascaux/en/).

At academic institutions, the face-to-face delivery of lectures has been enhanced, or replaced, by web-based learning management systems, such as WebCTTM and BlackboardTM.  PowerPointTM presentations are de rigueur at all kinds of venues – from the international scientific meeting to the high school classroom.   The use of images – whether in virtual or live classroom presentations – is considered essential, because we “learn more deeply from words with pictures than from words alone.”[3]

Caveats

The beauty of the Internet is the ability to make images  more readily accessible to a greater audience.  However there are caveats that users should be aware of before glibly downloading images from the Internet.

With sophisticated software tools, users can manipulate digital images.  They might add arrows to point out certain aspects of an image or enhance an image to improve its rendering.  However, manipulation of an image may misconstrue the original intent of the image or present fraudulent data, which leads to legal and ethical implications[4], i.e. infringement of copyright.  Digital images are protected by copyright, even if the image does not display the copyright symbol (©).  The Copyright Act of 1976 does not require the creator of a work to register or renew copyright.  Consequently anything fixed in a tangible medium – an e-mail, a voicemail message – is   protected by copyright. [5]  Since most digital images are copyrighted, users must get permission to use the images in presentations that do not meet the fair use criteria, or in instances in which the original work will be changed in any way. [6]

Internet search engines are critical for finding and retrieving images.  However, if the images do not have relevant descriptive tags or have not been indexed or cataloged using industry standards, retrieving relevant images can be time consuming.  Also, many institutions and organizations, such as Mayo Clinic, have not made their image repositories accessible from the Internet.  Reasons for not making them available vary – from an institution’s concern over copyright and intellectual property policies, to a lack of a peer-review process, and a perceived lack of recognition on the part of the institution for educational innovation. [7]  Mayo Clinic maintains its own institutional image repository, some of which is available from the Mayo intranet (http://mayoweb.mayo.edu/id-imagearchive/). According to Nancy Moltaji, Mayo’s media librarian, the Division of Media Support Services is in the process of converting their records to a web-based application which will allow Mayo employees the ability to search for images from the intranet.  Currently, there are over 45,000 illustrations, and 255,000 digital photographs in the Mayo media archive.

The technology with which we create and store images will continue to evolve.  Because we do not know if the systems or software programs that we’ll be using twenty-five years from now will be backward compatible, “…a long-range strategy is required to ensure that electronic images acquired today can be retrieved and viewed … in the future.”[8]   As a personal strategy, it is good practice to refresh images into new formats so that we can continue to access them in the future. 

Image-based websites

The Mayo Clinic Libraries website provides a resource page (http://library.mayo.edu/home/medical-images.html – Mayo Clinic only) of medically-related websites. The page is categorized by broad subject headings, e.g. Anatomy & Histology, Dermatology, General, Surgery, etc.   Some of the websites listed on this page offer many images yet they do not have search engines, which may seem a limiting factor for some.  Other sites are well indexed, with robust search engines and with the ability to store images   in personal folders.  Some require users to register with their sites although there is no fee involved.  Some sites allow users to download and use images for educational purposes only, some have public domain images which do not require copyright permission.  Some sites are peer-reviewed or rate the images, others do not.  The majority of sites request that users credit the website and/or author of the image. Here is a cross section of some of the medical image websites listed on the Library’s website:

Atlas of Ophthalmology <http://www.atlasophthalmology.com/bin/atlas?lang=en> This online atlas, endorsed by the International Council of Ophthalmology, aims to provide education to students, residents and specialists by  “showing and describing the various stages of the clinical picture as it is seen in photographs and by ancillary tests…”  Users have permission to freely use the images, of which there are approximately 3000.  They are categorized hierarchically by anatomic site, then by disease states.  There is a basic search function.  Each record has an ID number as well as the ICD10-Code, a diagnosis, comment and keywords. The images are not peer-reviewed.

The Big Picture Book of Viruses <http://www.virology.net/Big_Virology/BVHomePage.html> This site, from Tulane University, is a portal to images stored on other websites.  Its aim is to help educate those interested in learning more about viruses.  There is no search engine.  Images can be located by name (virus families” or individual viruses), by structure or genome, by host or by disease.  The description includes the taxonomy, host, genome, and morphology.  The images are in thumbnails, with information as to who submitted the image.  Users would need to contact the author of the image for copyright permission.

DermAtlas <http://dermatlas.med.jhmi.edu/derm/> DermAtlas is hosted by Johns Hopkins University.  Over 8300 dermatologic photographs and histologic slides have been made available for teaching purposes.  Images are indexed by categories, diagnoses and body sites, and tagged with metada allowing for advanced searches. Images are also linked to references in PubMed, eMedicine and several dermatology journals.  Copyright is held jointly by the contributing author and DermAtlas. Storing images from the DermAtlas on other web servers is not permitted.

Google Images <http://images.google.com/> According to its website, Google has billions of indexed images (http://www.google.com/help/faq_images.html).   A searcher can limit a search by file type (jpg, gif, pnf), by file size (small, medium, large) or by domain.  Google Images does capture images that have been published in biomedical and scientific journals, which most of the other sites do not.  It is up to the searcher to determine if the images are copyrighted and to seek permission from the copyright holder.

HEAL (Health Education Assets Library) <http://www.healcentral.org> HEAL is a national digital library, whose aim is to provide free access to multimedia files for all levels of health sciences education.  Initially funded by the National Science Foundation in 2000, HEAL received additional funding from the National Library of Medicine in 2003.  Individuals can submit multimedia files which undergo a peer-review process.  HEAL has partnered with other institutions to make their digital libraries available.  One can search across multiple collections. Records are indexed using Medical Subject Headings (MeSH).  The HEAL developers also use a metadata schema based on the Educause IMS (Instructional Management System) Standard.  Users must register in order to view and download records.  They can save images to a folder and download the entire contents to a zip or gzip file.

Historical Anatomies on the Web from the National Library of Medicine <http://www.nlm.nih.gov/exhibition/historicalanatomies/ home.html> This is a digital project from the NLM.   Selected images from several important texts from the Library’s historical anatomical atlases collection have been scanned in at high resolution.  The works are arranged alphabetically by author and can be browsed by title.  Each title is linked to the bibliographic information in the Library’s online catalog.  All the images are in the public domain however NLM does want users to acknowledge credit to the Library when using these materials.

HONMedia from Health on the NET Foundation <http://wolfgang.hcuge.ch/Media/media.html> HONMedia acts as a search engine for locating medical images; it is not a repository of digital images.  For example, the images in DermAtlas (noted earlier) are linked from HONMedia.  MeSH classification is used to index images.

Karolinska Institute <http://www.mic.ki.se/MEDIMAGES.html> The Karolinksa Institute in Stockholm, Sweden is one of the portal websites.  It has compiled a comprehensive list of websites that link to biomedically related images and illustrations.  The list is categorized by broad MeSH subject headings, e.g., “cardiovascular system”, “endocrinology,” however, one cannot search for images from this site.

National Eye Institute <http://www.nei.nih.gov/photo/> The National Eye Institute has made its collection of photographs, illustrations and animations available from its website.   Searcher can do keyword searchers, or browse by topic.  The files are not copyrighted, but users are asked to credit to the NEI or the NIH.

PEIR (Pathology Education Instructional Resource) Digital Library <http://peir2.path.uab.edu/pdl/dbra.cgi?uid=default&vie w_search=1> The PEIR Digital Library is made available by the University of Alabama, Birmingham, (UAB) Department of Pathology for use in medical education. Originally, the resource only included digitized 35mm pathology slides; it has evolved into a digital library of  images, test items and tutorials.   The images from UAB’s Department of Radiology, the Armed Forces Institute of Pathology and the Slice of Life have been included.  Users can search by specific source, by type (cytology, microscopic, gross, diagram, etc) and then by keyword.  Each record has a thumbnail, with metadata tags of type, organ/system and description. There are buttons to zoom in or out, and to save images to a “shopping cart.”  Although access is unrestricted, images can only be used for non-profit educational activities.

Public Health Image Library from the Centers for Disease Control <http://phil.cdc.gov/Phil/home.asp> This site is sponsored by the National Center for Health Marketing, the National Center for Infectious Diseases and the Information Resources Management Office.   It includes photographs, illustrations and multimedia files germane to the field of public health.  The files are indexed using MeSH, and organized into hierarchical categories of people, places and science.  As this is a federal government site, the majority of images are in the public domain.

Sci7 Biomedical Images <http://www.biomedimages.com/> This site, a project of Sci7 (http://www.sci7.com/), has over 22,000 life sciences images.  They are available under the terms of the Biomed Central open access license, meaning that users are free to use the work as long as they cite the original author.  The images appear to have been originally published in journal articles. There is only a basic search engine.

Ann Farrell, Librarian, Winn Dixie Medical Library Mayo Clinic Florida

REFERENCES

  1. LaPorte RE, Linkov F, Villasenor T, et al. Papyrus to PowerPoint (P 2 P): metamorphosis of scientific communication. BMJ 2002;325(7378):1478-81.
  2. Tsafrir J, Ohry A. Medical illustration: from caves to cyberspace. Health Information & Libraries Journal 2001;18(2):99-109.
  3. Creating Passionate Users. 2005. (Accessed March  6, 2006, at http://headrush.typepad.com/creating_passionate_users/2005/10/words_pictures_.html.)
  4. Suvarna SK, Ansary MA. Histopathology and the ‘third great lie’. When is an image not a scientifically authentic image? Histopathology 2001;39(5):441-6.
  5. Gordon-Murname L. Generosity and copyright: creative commons and creative commons search tools. Searcher 2005;13(7):16-23.
  6. Hoffmann GM.  Copyright in cyberspace 2: questions and answers for librarians. New York: Neal-Schuman Publishers; 2005.
  7. Uijtdehaage SH, Contini J, Candler CS, Dennis SE.  Sharing digital teaching resources: breaking down barriers by addressing the concerns of faculty members. Academic Medicine 2003;78(3):286-94.
  8. Blado ME.  Management of the picture archiving and communications system archive at Texas Children’s Hospital. Journal of Digital Imaging 2001;14(2 Suppl 1):84-8.

(Originally published in e-knowledge.net: Quarterly Newsletter of the Mayo Clinic Libraries, Winter Issue No. 21, March 2006).

Entry filed under: From the Archives.

NLM Update Citation Indexes: More than Just Web of Science

2 Comments

  • 1. tamas  |  June 3, 2009 at 3:53 am

    Very useful article. Unfortunately, this link does not work: http://library.mayo.edu/home/medical-images.html

  • 2. mlrethlefsen  |  June 3, 2009 at 7:09 am

    Thanks for your comment, Tamas–we will update the article to reflect that the URL in question is for Mayo Clinic only.



%d bloggers like this: