Secondary Logo

Journal Logo

Special Theme: Education Technology: Special Theme Articles

Introducing HEAL

The Health Education Assets Library

Candler, Chris S. MD; Uijtdehaage, Sebastian H. J. PhD; Dennis, Sharon E. MLS

Author Information
  • Free


The advent of the personal computer has served as a change agent for many disciplines and has created many new possibilities for medical educators. The use of educational technology in medical education has blossomed over the past decade, driven by truly remarkable advances in information technology.1 In today's classroom, instructors increasingly turn to digital resources such as multimedia images, videos, animations, and simulations to portray both basic and clinical science concepts with unparalleled clarity and realism. In an attempt to understand how these trends might influence the future of medical education, the Association of American Medical Colleges (AAMC) recently launched a forward-looking Delphi study to explore future uses of the Internet and information technologies at academic medical centers.2 Academic leaders were asked to speculate on the use of technology in education and other areas. The resulting data highlighted the growing importance of technology in educational delivery and assessment. Other groups have echoed these ideas and have gone so far as to suggest that most medical schools will ultimately move substantial portions of their curricula online.3

Throughout the past two decades several precocious medical schools strategically invested in educational technology using proprietary techniques that could not be easily transferred to other schools.4 In most cases, medical schools and faculty did not develop educational resources with the intent to share with other schools. There were several notable exceptions, however, such as the Slice of Life videodisc, which offers over 40,000 images to educators.5 With the watershed emergence of the World Wide Web, resources developed at one medical school could easily be shared with students and faculty at medical schools across the globe. Today, medical students now share Web sites as they once shared class notes. Yet even since the advent of the Internet, there remains a digital divide between those schools that have made educational technology a strategic priority and those schools that lack a commitment to instructional technology or that simply lack the resources to implement their vision.6 Fortunately, the Internet has the potential to connect the digital haves with the digital have-nots, as evidenced by many high-profile medical education Web sites freely available to both students and faculty.


The measureless and interconnected nature of the Internet, for all its benefits, can also be a double-edged sword to educators trying to identify educational multimedia resources for use in the classroom. On the one hand, the Internet has the potential to provide direct and worldwide access to valuable educational resources. On the other hand, finding high-quality and suitable multimedia requires careful scrutiny of a bewildering number of documents and multimedia items; separating the wheat from the chaff is often not worth the time. Unfortunately, most Web-based search utilities overwhelm educators with useless materials and are not able to limit searches to high-quality medical education multimedia. Furthermore, finding materials within an institution is often equally challenging. As we pointed out previously,7 many medical schools' Web-based curricula suffer from insufficient planning and inconsistent use of user-friendly authoring programs that ultimately create chaotic tangles of resources that cannot be searched effectively, or used for new purposes efficiently, or updated consistently. Not uncommonly, the number of HTML and image files on a curriculum Web server may exceed 10,000 items, many of which lack descriptive information and proper indexing. Ironically, faculty are often frustrated with the inability to find not only resources created by others but also their own because of the “intangible” nature of digital materials. In the past, faculty could organize “tangible” media, such as slides, by physically labeling them or organizing them in carousels. Digital multimedia, however, can easily become lost within myriad files, folders, and servers if they are not properly stored and indexed.

Because useable materials are difficult to obtain, faculty members often create their own multimedia or have them created on their behalf. Given that the contents of curricula across medical colleges are comparable, if not identical to some extent, faculty tend to reinvent the wheel when they create a diagram or develop a case. A radiograph of a lung tumor, an animation of an electrocardiogram, a labeled sagittal cross-section of the brain are all examples of multimedia that each institution is likely to employ. Creating interactive multimedia, however, is costly. For example, creating an interactive animation depicting the hemodynamics and electrophysiology of the heart (e.g., Hyperheart8) requires a lengthy process of instructional and graphic design, programming, and evaluation. Such resources can easily cost over $10,000 apiece to design and develop when faculty and staff time are considered. Ironically, many materials are freely available in a similar form at other medical schools. Clearly, educators cannot individually or collectively afford to reinvent this wheel, considering the boundless collaboration opportunities afforded by the Internet.

Duplicating the creation of costly multimedia could be avoided in part by sharing teaching resources with colleagues. Unfortunately, sharing of multimedia within and among institutions is thwarted by both technical and non-technical obstacles. First, there is no widely recognized platform for sharing multimedia with a national audience. A national repository of medical educational resources to which faculty can contribute their multimedia does not exist. One could make a resource publicly available by embedding it in a Web page, but the chances of reaching an audience of medical educators are severely limited by of the vast number of indexed Web pages available on the Internet (approaching two billion by some estimates9). Secondly, most institutional Web-based curricula are based on inefficient HTML file-based architectures that inextricably couple content storage and content delivery functions. Because this online educational content is hopelessly intertwined within the delivery system, these resources are essentially buried and cannot be easily discovered or shared as individual items independent of the educational contexts in which they are used (cases, tutorials, lecture notes, etc.). Many medical schools are turning to content management systems built upon sophisticated databases that structure and index their online resources.10

Another major barrier that severely limits the portability of teaching resources is the lack of a universal method for cataloging medical education multimedia. Consistent assignment of descriptive information (metadata) using a controlled vocabulary is a necessary condition for building educational repositories and for effective repurposing and exchanging their contents. The IMS Learning Resource Meta-Data Specification11 is a globally supported standard that defines a mechanism to assign educational materials with descriptive metadata. This standard, however, is necessarily generic and alone is not sufficient to capture the information that may be important to medical educators, such as diagnosis, organ system, and disease category.

Some of the obstacles for sharing of multimedia are non-technical in nature. A recent focus group study among medical faculty12 revealed that confusion exists about the extent to which current copyright laws protect against unlawful use of the multimedia. Many fear that by making the resources available on the Internet, others may utilize them without appropriate attribution, or worse, claim ownership illegally. Adding to the confusion are ill-communicated institutional rules to determine who owns the copyright of the multimedia in the first place. For instance, to some faculty it is unclear whether or not they can legally grant user licenses to colleagues for using their materials that they developed. Until ownership issues are clarified, many faculty are not willing to share their resources with colleagues—let alone with the public at large.


To address these issues a consortium of medical schools across the United States met in 1998 to explore the feasibility of sharing multimedia resources among medical colleges.* Initially, a consensus was achieved on how health sciences multimedia assets should be cataloged. This consensus was a first and important step towards easier sharing of digital multimedia among institutions. While working with the consortium, we formed a working group to develop metadata standards designed to facilitate the exchange of educational content and explore the creation of a national repository. We had previous experience designing and implementing two multimedia repository projects: the University of Utah Spencer S. Eccles Multimedia Catalog and the UCLA Multimedia Database.13–17 Based on these experiences and the desire to develop a more global solution, we ultimately agreed that external funding was necessary to continue work on a national multimedia repository. In 2000, we received funding from the National Science Foundation's Digital Library Initiative. The two-year NSF funding allowed for the creation of the Health Education Assets Library (HEAL) project.18 As part of this NSF initiative, HEAL will become a component of the National Digital Library for Science, Mathematics, and Technology Education.19 In 2002, the HEAL team received an additional $780,000 from the same NSF program to continue the development of HEAL. Lastly, HEAL has formed alliances with the National Library of Medicine (NLM) and the AAMC.

HEAL's mission is to improve access to teaching and learning resources for health science educators and students across the United States; to promote sharing of teaching resources; to foster interoperability of resources; and to provide a reliable and enduring infrastructure. This will be accomplished through the following activities:

  • ▪ Sharing the HEAL metadata schema for health educational materials in a way that is compatible with international standards such as IMS standards
  • ▪ Developing a core collection of high-quality teaching resources for the health sciences
  • ▪ Developing clear intellectual property and patient confidentiality policies
  • ▪ Implementing an architecture that incorporates additional collections and allows other organizations to tap directly into HEAL services such as searching and browsing
  • ▪ Studying and identifying obstacles to collection development
  • ▪ Providing a sustainable quality assurance mechanism

HEAL is not designed to be a course delivery system, a standardized curriculum, or a substitute for an institution's course or curriculum. Alternatively, HEAL provides building-block multimedia items such as images, videos, and animations, and textual materials such as cases and quiz questions. Educators may incorporate these materials into their own teaching resources that are customized to reflect the local needs and teaching styles. These might include (1) online curriculum materials such as atlases, multimedia textbooks, interactive syllabi etc., (2) PowerPoint presentations, (3) problem-based learning cases, (4) practice quizzes, and (5) online assessment programs.

Instructors may use HEAL to freely browse, search, and download multimedia from many health sciences multimedia collections through HEAL's central search page. Both a simple and an advanced search page will be available. After performing a search the user is presented with a search-results page that includes basic metadata information about the multimedia item and a small “thumbnail” reference image. The user may view the full-size image, play the full audio, video, or animation, and examine the complete metadata for any given record. Once the desired resources have been selected the user may download all the items and the descriptive metadata in one simple step. Each resource contains a clear description of copyright limitations, so that faculty may understand how the material may be used.

HEAL content will derive from both individual faculty contributions and relationships with large institutional collections. Faculty contributors may easily donate individual resources using HEAL's upload interface, where they are asked to provide a basic set of metadata (including title, description, author, and copyright information). Once uploaded, the item is subject to approval by the HEAL team, which will ensure that the content is relevant to health care education. After approval, a professional librarian will catalog the item. Cataloging may include the addition of National Library of Medicine Medical Subject Headings (MeSH) or other pertinent controlled vocabularies.


Perhaps the most compelling feature of HEAL is its potential to unite individual institutional collections into one virtual collection that is searchable through a single index. In addition to the core collection, HEAL users will be able to access high-quality multimedia collections hosted by affiliated institutions. This “federation” of affiliated collections will ultimately form a vast, distributed repository available to medical educators across the globe. The value of HEAL grows with each partner collection that joins the HEAL federation, providing instructors with a tremendous number and variety of resources available for download. As Figure 1 illustrates, many institutional collections may connect to the HEAL federation. Both institutions that have an existing content management system and institutions that do not may join the HEAL federation. Institutions that do not have a content management system may freely obtain the HEAL software and implement it on their servers. The distributed nature of this system will enable access to a greater number of multimedia resources than would be possible with a centralized system. In addition, the HEAL system will be connected to the National Sciences Digital Library, which will further amplify the breadth and depth of the system.

Figure 1:
The HEAL Federation will incorporate collections from many medical schools, all searchable via a central HEAL server. Future applications may tap directly into HEAL metadata services such as searching and browsing.

While the master HEAL database will contain many thousands of multimedia items, it is not essential that all content be stored centrally. In fact, partner collections may continue to host their own content as long as they allow the HEAL system to periodically “harvest” their metadata so that all such content is accurately reflected in the central HEAL index. To further extend the reach and utility of the central index, other organizations may tap directly into HEAL services such as searching and browsing. It is important to note that even after they join the HEAL system, individual schools' collections will always be clearly branded with the institutions' logos. Thus, the source institution and contributor will always be clearly evident for each and every multimedia object indexed by the HEAL system.

In order to create a repository that allows medical school faculty to share multimedia with other schools, the HEAL system is designed according to internationally recognized standards. In particular, the HEAL federation is built upon network standards defined and promoted through the National Sciences Digital Library (NSDL) initiative. Other HEAL standards are based on and are compatible with the internationally-recognized IMS metadata and content packaging standards, specifications developed to promote interoperability of applications and services in distributed learning. EXtensible Markup Language (XML), provides the basis for the above standards.20 In addition, HEAL will incorporate controlled vocabulary standards to describe the multimedia resources. Initially, the items will be cataloged using MeSH. Other controlled vocabularies, such as the Unified Medical Language System (UMLS), are also being investigated for incorporation into HEAL.

Adoption of the HEAL metadata standard offers several advantages to educators. For instance, medical schools can create and maintain their own multimedia collections or portions thereof; as long as the metadata conforms to the HEAL standard these collections may be included in HEAL's master index. Secondly, resources downloaded from the HEAL repository can be readily included in other course delivery systems, such as WebCT,21 that support the IMS content packaging standard.

HEAL is being developed in an open-source spirit and will share all its applications, standards, and content freely with educational institutions. Only open-source and platform-independent technologies are being employed, such as Java Server Pages™ (JSP), Java Beans™, and XML. The advantage of using nonproprietary technologies is that external institutions may readily customize and extend the core HEAL application to accommodate their local needs.


The first major phase of the HEAL application has been completed and the core content collection is indexed and available for download. This includes several thousand digital images, videos, and animations covering a wide scope of topics and disciplines such as anatomy, pathology, neurology, and radiology. This is available at 〈〉. Ultimately, the HEAL team anticipates that the number of available resources will grow to many tens of thousands of items. Currently, the HEAL team is working to link several major collections to HEAL, which should significantly magnify the resource base. The HEAL database may also be customized to accommodate future types of educational resources.

Planned enhancements to the HEAL system include a peer-review function, as HEAL has the potential to provide faculty highly visible recognition via their contributions to a large-scale educational community. Such recognition may be important considering the many hours it takes to create truly outstanding educational resources. In the near future HEAL may serve this peer-review function by allowing the educational community an opportunity to comment on the accuracy, quality, and relevance of each resource. This feature may help individuals validate their efforts as scholarly products to be considered in promotion and tenure decisions.

The technology pundit Stewart Brand is credited with the assertion that “information wants to be free.”22 Fundamentally, the purpose of HEAL is to promote this notion by forging a large-scale education community based on the generosity of individuals and institutions willing to share their materials with others. Faculty and institutions no longer need to expend resources duplicating educational materials or tracking down high-quality educational multimedia, as this system will provide instructors with free access to a vast distributed repository of high-quality educational materials that may be downloaded and used within any course or curriculum. Furthermore, participation in the HEAL federation will benefit contributors, as the combined resources, services, and reach of the entire system will add value to not only the smallest contribution but also entire collections. By connecting contributors and consumers, HEAL can improve access to teaching and learning resources for all health science educators and thereby create a new kind of educational community.


1. MacKenzie JD, Greenes RA. The World Wide Web: redefining medical education. JAMA. 1997;278:1785–6.
2. The [email protected] Delphi Study: 〈〉. Accessed 11/11/02.
3. The Blue Ridge Academic Health Group: Report 5. Charlottesville, VA: BRAHG, 2001: 17.
4. Dev P, Hoffer EP, Barnett GO. Chapter 17: computers in medical education. In: Shortliffe EH, Perreault LE (eds). Medical Informatics: Computer Applications in Health Care and Biomedicine. 2nd ed. New York: Springer, 2001: 613.
5. Stensaas SS. Animating the curriculum: integrating multimedia into teaching. Bull Med Library Assoc. 1994;82(2):133–9.
6. Moberg TF, Whitcomb ME. Educational technology to facilitate medical students' learning: Background Paper 2 of Medical School Objectives Project. Acad Med. 1999;74:1145–50.
7. Candler CS, Andrews MD. Avoiding the great train wreck: standardizing the architecture for online curricula. Acad Med. 1999;74:1091–5.
8. Hyperheart. 〈〉 Accessed 11/11/02.
9. Google Internet site: 〈〉. Accessed 11/11/02.
10. Lee MY, Albright SA, Alkasab T, Damassa DA, Wang PJ, Eaton EK. Tufts Health Sciences Database: lessons, issues, and opportunities. Acad Med. 2003;78:254–64.
11. IMS Global Learning Consortium Web Site, 〈〉 Accessed 11/11/02.
12. Uijtdehaage SHJ, Contini J, Candler CS, Dennis SE. Sharing digital teaching resources: breaking down barriers by addressing the concerns of faculty members. Acad Med. 2003;78:286–94.
13. Uijtdehaage S, Dennis S, Candler C, Stensaas S. Cataloguing and exchanging digital multimedia. Paper presented at the Slice of Life 2000 Conference, Salt Lake City, UT, June 27–July 1, 2000.
14. Uijtdehaage S, Kleerup E, Kumpart E, Dennis S, Candler C. An online multimedia database for faculty use. Paper delivered at the Western Group on Educational Affairs, Asolimar, CA, April 24–26, 2000.
15. Uijtdehaage S, Kleerup E. Development of a multimedia database for faculty use. Presentation given at the symposium “Instructional Technology in Medical Education,” UCLA, March 18, 2000. Published at 〈〉. Accessed 11/11/02.
16. Uijtdehaage S, Relan A. Designing Web-based databases for effective medical education. Paper presented at Western Group on Educational Affairs, Park City, UT, April 10–12, 1999.
17. Uijtdehaage S, Relan A. Designing Web-based image databases to enhance medical education. Paper presented at AACE WebNet99 Conference, Honolulu, HI, October 25–30, 1999.
18. The Health Education Assets Library 〈〉. Accessed 11/11/02.
19. Science, Mathematics, Engineering and Technology Education Digital Library 〈〉. Accessed 11/11/02.
20. The XML Cover Pages, XML Metadata Specification, 〈〉. Accessed 11/11/02.
21. WebCT Learning Management System〈〉. Accessed 11/11/02.
22. Brand S. The Media Lab: Inventing the Future at MIT. New York: Viking Penguin, 1987.

*The Consortium participants represented the University of California Los Angeles, University of California Davis, University of California San Francisco, University of California San Diego, Drew University, University of Southern California, Loma Linda University, Stanford University, University of Nevada, University of Oklahoma, and University of Utah.
Cited Here

© 2003 Association of American Medical Colleges