HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Search for Related Content PubMed PubMed Citation Articles by McGreevy P Articles by Miller N

OLIVER: An Online Library of Images for Veterinary Education and Research

	ABSTRACT

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 
As part of a strategic move by the University of Sydney toward increased flexibility in learning, the Faculty of Veterinary Science undertook a number of developments involving Web-based teaching and assessment. OLIVER underpins them by providing a rich, durable repository for learning objects. To integrate Web-based learning, case studies, and didactic presentations for veterinary and animal science students, we established an online library of images and other learning objects for use by academics in the Faculties of Veterinary Science and Agriculture. The objectives of OLIVER were to maximize the use of the faculty's teaching resources by providing a stable archiving facility for graphic images and other multimedia learning objects that allows flexible and precise searching, integrating indexing standards, thesauri, pull-down lists of preferred terms, and linking of objects within cases. OLIVER offers a portable and expandable Web-based shell that facilitates ongoing storage of learning objects in a range of media. Learning objects can be downloaded in common, standardized formats so that they can be easily imported for use in a range of applications, including Microsoft PowerPoint, WebCT, and Microsoft Word. OLIVER now contains more than 9,000 images relating to many facets of veterinary science; these are annotated and supported by search engines that allow rapid access to both images and relevant information. The Web site is easily updated and adapted as required.

Key Words: learning objects • image bank • indexing • illustrations • videos

	PROJECT OVERVIEW

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 
As part of the strategic move by the University of Sydney toward increased flexibility in learning, the Faculty of Veterinary Science (FVS) has undertaken a number of developments involving Web-based teaching and assessment. This is a trend reflected in educational institutions worldwide.^1–8 These include self-directed learning using Web-based case studies, formative assessment using WebMCQ,¹ and the application of Microsoft PowerPoint² to enhance didactic and interactive presentations. Current projects also include the development of two Web-based courses to be mounted on WebCT³ and the investigation of flexible learning methods within the new curriculum for the Bachelor of Veterinary Science degree offered from 2000.

Teaching staff in the FVS hold, or have access to, a large number of images and graphics relevant to their research, teaching, and consultancy programs. These images include, for example, animals in health and disease, husbandry techniques, radiographs, surgical methods, and graphs of experimental data. The teaching staff are also skilled in producing learning objects to demonstrate best practice of standard techniques in animal husbandry, medicine, and surgery.

A number of challenges have been encountered by our faculty and others when images have previously been used for online teaching purposes,^{3, 8} including the following:

• Difficulty in storing, cataloguing, and retrieving images
  
• Difficulty in sharing image resources, both between faculty members and between faculty and students
  
• Frequent use of multiple copies of images in multiple online learning resources, making tracking of changes and image updating extremely difficult
  

In order to improve and integrate Web-based learning, case studies, and didactic presentations for veterinary and animal-science students, we propose to establish an online library of images and other learning objects for use by academics in the Faculties of Veterinary Science and Agriculture. The Online Library of Images for Veterinary Education and Research (OLIVER) contains images relating to many facets of veterinary science, annotated and supported by search engines that allow the rapid access to both images and relevant information. OLIVER can be used as a learning resource in its own right, or images from the database can easily be linked directly into other Web-based teaching resources.

	PROJECT OBJECTIVES

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 

1. To maximize the use of the FSV's teaching resources by providing a stable archiving facility for graphic images and other multimedia learning objects.
   
2. To create a database structure that allows flexible and precise searching, integrating indexing standards, thesauri, pull-down lists of preferred terms, and linking of objects within cases.
   
3. To archive learning objects in common, standardized formats so that they can be easily directly imported for use in a range of applications, including Microsoft PowerPoint, WebCT, and Microsoft Word.
   
4. To create a user-friendly interface with a professional appearance that enables users with a range of technical abilities to easily search for and make use of learning objects for teaching purposes.
   

	DATABASE FUNCTIONALITY

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 
Four main functions have been designed for the database to meet the objectives of the project:

1. Creator/Management functions via an interface that
   
   • allows easy uploading of learning objects from users’ available network and local drives.
     
   • conforms objects to an agreed standard, including file format, name, resolution and size.
     
   • links series of records into groups.
     
   • requires the creator to complete all mandatory fields in a record.
     
   • requires the creator to submit the incomplete record to a quality controller/indexer.
     
   • requires the quality controller/indexer to add structured indexing terms and check entries for errors before releasing them to the database.
     
   
   
2. Basic and advanced searching functions via an interface that
   
   • allows for full-text searching across the database, using a Google-like syntax that allows for nested Boolean search terms.
     
   • searches the database using SNOMED⁹ thesaurus terms.
     
   • returns results based on relevance to the search terms.
     
   • displays Help information and links to example searches and FAQs.
     
   
   
3. Display functions via an interface that
   
   • allows the user to easily redo, narrow, or expand the search if the hit rate is too high or too low or if records retrieved are irrelevant.
     
   • displays the total number of records retrieved by the search and sequentially numbers the records retrieved.
     
   • displays a brief description of the record with a hyperlink to the full record and links to associated images in the set or case.
     
   • displays a thumbnail image for each record retrieved.
     
   
   
4. Output of images for use in other teaching resources:
   
   • They system automatically displays an image optimized for use in Microsoft PowerPoint presentations or for display via the Internet.
     
   • The system outputs images of preset size for display in a remote Web-based resource by embedding a URL within the HTML code of the remote site.
     
   • A pluggable authentication mechanism ensures that OLIVER is compatible with the university's internally developed single-sign-on system, as well as allowing accounts to be created within OLIVER, giving the faculty flexibility in who it allows into OLIVER while retaining ease of use for students.
     
   
   

	SYSTEM ARCHITECTURE AND COMPATIBILITY

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 
OLIVER is built using server-side Java. To ensure the best cross-platform compatibility, no part of the site's functionality depends on any client-side functionality other than HTML 4 and cookies. This ensures that clients can access the site using any Web browser on any platform.

A number of open-source components are used in OLIVER:

• Tomcat (a Java Servlet container)⁴
  
• Lucene (a free text search engine)⁵
  
• ImageMagick (image-manipulation software)⁶
  

Using these free tools cut down the development time of OLIVER enormously, allowing the developers to concentrate on writing the software to manage the images and their metadata.

The server is running Red Hat Linux⁷ with the Sun Java runtime environment.⁸ OLIVER can be used with a range of browsers, including Netscape Navigator and Internet Explorer, and is compatible with Windows-, Linux-, and Macintosh-based systems.

	USING OLIVER

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 
On login, users are identified as either staff or students, and this dichotomy is used to determine the privileges that are granted within OLIVER by masking certain buttons and, thus, certain functions.

Uploading Objects and Creating Records
Uploading learning objects and creating records is a function currently restricted to FVS staff.

Objects can be added into OLIVER using any Web-enabled computer. OLIVER can manage files of any type, including images (.jpeg.tif.bmp), video (.mov.mpeg), PDF, and MS Word documents. However, only some image formats can be thumbnailed and resized. Other formats cannot be transformed in any way by OLIVER at the time of writing.

When uploading a new learning object, users must complete all mandatory metadata fields and as many other fields as possible (see Table 1 for full listing and description of metadata fields).

Users are advised to upload the highest-resolution image available, to ensure that the images do not become obsolete as screen and printing resolutions increase over the years.

Indexing of Objects
Images entered into OLIVER by FVS staff are automatically withheld from other users until they have been professionally indexed. This ensures the quality of the metadata and the integrity of the database.

An indexer ensures that there is an accurate description of the object and classifies each image against two thesauri, Medical Subject Headings (MeSH)¹⁰ and SNOMED.⁹ The SNOMED classification has been built into OLIVER, allowing a full search of the classification system and simple tagging of SNOMED terms to each object.

Using Groups
Objects that are linked in some way can be clustered together by faculty-level users into either general groups or clinical groups.

Each individual image within a group still requires its own metadata and can be retrieved separately. Both new and existing images can be added to any group, and images can be ordered within any given group. An image can be in more than one group.

General groups are assigned a title, a description, and subject headings. These groups are generally used by faculty members to cluster objects together for purposes such as specific tutorials or lectures.

Clinical groups are assigned a title, a description, case-management notes, a case history, subject headings, clinical findings, and a diagnosis. Clinical groups are designed to enable the clustering of a set of objects around a particular case (such as interoperative, pathological, and radiological images from the same case).

Searching for Objects
After logging in, students and faculty can use OLIVER to search for and download learning objects such as images and videos. A search can be conducted for individual images and groups of images. The 20 most frequently accessed images from OLIVER are displayed on the front page.

The basic output unit of the database is termed a "record." Hits from a search are displayed as thumbnails for each record. Records can be viewed by clicking on the thumbnails. Associated metadata are displayed with each record.

Downloading Objects
Several preset image sizes are available for download. These include image sizes optimized for use in Microsoft PowerPoint, for Web delivery, and for printing. When an image size is selected, an version of the image with the required resolution is produced on the fly and then cached for subsequent downloads. All "Web Page," "PowerPoint," and "Full Size" (i.e., print-resolution) images are in JPEG format. An additional option allows the user to select the desired resolution by typing it in (in pixels per inch, or PPI), up to the maximum available.

Editing Images
An edit function is available on OLIVER. Faculty members first use the Search function to find the images they wish to edit or delete. Clicking "Edit" on the Record page allows the user to amend the text associated with any image or group, as well as the image or group itself. The system will not allow anyone other than the original submitter and the OLIVER management team to edit any contribution to OLIVER.

Providing Feedback
By clicking "Feedback," students and faculty can contribute comments or questions on any technical, search, or content issue. Feedback on individual images or the text associated with them can be given by clicking "Feedback on This Image," available with every record.

	OLIVER SHOWCASE

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 
The OLIVER site contains a password-free showcase section where visitors can view commonly used images and screen grabs detailing how searching and uploading of images is performed within OLIVER. The showcase can be accessed at <http://oliver.vetsci.usyd.edu.au>.

	DISCUSSION

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 
With over 9,000 indexed records, OLIVER is meeting its objectives, in that it is a learning resource in its own right and can easily be linked directly into other Web-based teaching resources. The size, functionality, and aims of OLIVER match those of a number of substantial international image databases. ^{1, 2} Numerous image donations have been made to OLIVER by distinguished retiring veterinarians, alumni, and government agencies. The project has effectively maximized the use of the faculty teaching resources by providing a stable repository that allows flexible and precise searching and linking of objects within cases. Most users who provide feedback comment on how easily OLIVER's content can be imported for use in a range of applications, including Microsoft PowerPoint, WebCT, and Microsoft Word.

While current undergraduate students and alumni regularly donate images to the FVS for use in OLIVER, the addition of new records relies on the commitment of teaching staff conversant with the veterinary significance of the learning object to upload images with sufficient detail to make indexing thorough. With its unique wildlife and range of veterinary pathologies, Australia offers veterinarians a fascinating challenge that is reflected in OLIVER's content. To complete the coverage of veterinary topics in a global sense, collaboration with overseas veterinary education providers (especially from the Northern hemisphere) is being explored.

The development team believes that the greatest challenges to OLIVER are issues of sustainability and quality-control management. To be sustainable, the project must be self funding. The chief expenses incurred by OLIVER relate to technical maintenance, indexing of records, and annual license fees for SNOMED. Making OLIVER financially sustainable is likely to involve exploring central university funding, the commercial use of images from the collection, and corporate sponsorship.

The use of professional indexers has greatly improved the overall quality of records, resulting in highly retrievable images and few erroneous search returns. However, this high level of quality is offset by the cost of professional indexing and the increased complexity of inputting images for individual contributors. The OLIVER development team has recognized that a balance must be struck between quantity in the short term and quality in the long term.

The University of Sydney is now developing new digital repository software called iSpheres, which is based on what we have learned from OLIVER as well as from other repositories on campus. Image repositories such as OLIVER have shown that there is a clear need for discipline-specific repositories, and iSpheres is designed to satisfy that need. It is also built in Java and is designed for distributed sharing of images and to allow the building of customized user interfaces to repositories for research and learning. The primary interface to an iSphere is via SOAP (Simple Object Access Protocol), which allows Web interfaces to be built in most modern programming languages and platforms and iSpheres interfaces to be integrated into most learning-management systems and existing Web interfaces. iSpheres also uses a flexible-connector back end for its data storage, allowing developers to build connectors into virtually any data source.

Proposed improvements to OLIVER include the following:

• Implementing creator functions for students, both to increase the number of images in the database and to increase the use of the resource as a teaching and learning tool in its own right
  
• Enabling seamless links with other veterinary science image banks
  
• Developing a multiple-user interface that displays only those data relevant to each user
  
• Encouraging the use of OLIVER by research staff to store images, to improve sustainability and increase the number of images
  

View larger version (41K): [in this window] [in a new window]   Figure 1: The OLIVER home page.

View larger version (61K): [in this window] [in a new window]   Figure 2: An example of the results of a simple search in OLIVER; the left sidebar shows hits as thumbnails.

View larger version (64K): [in this window] [in a new window]   Figure 3: An example of an individual record in OLIVER; the left sidebar allows downloading according to specific uses or customized requirements.

	Acknowledgments

	Footnotes

 
AUTHOR INFORMATION

Paul McGreevy, BVSc, PhD, is Associate Professor in Animal Behaviour in the Faculty of Veterinary Science (B19), University of Sydney, NSW 2006 Australia. E-mail: paulm{at}vetsci.usyd.edu.au. His research interests focus on stereotypic behaviors in various species, measures of welfare in captive and domestic species, and the husbandry and training of companion animals. His innovations in teaching focus on the use of information technologies to support flexible delivery of learning opportunities.

Tim Shaw, BSc (Hons1), PhD is the Director of the Centre for Innovation in Professional Health Education, Faculty of Medicine, University of Sydney, NSW 2006 Australia. E-mail: tims{at}ciphe.med.usyd.edu.au. His research interests lie in the application of flexible learning in the workplace.

Daniel Burn, BSc (Hons), is the IT Development Manager in the Faculty of Medicine, University of Sydney, NSW 2006 Australia. E-mail: daniel{at}dme.usyd.edu.au. His software-development interests lie in the use of multimedia materials in e-learning, and he has been involved in the creation of a wide range of e-learning systems used within the Faculty of Medicine and elsewhere.

Nick Miller, BIT, is a Senior Software Developer in the Faculty of Medicine, University of Sydney, NSW 2006 Australia. E-mail: nmiller{at}med.usyd.edu.au. His software-development interests lie in developing e-learning systems using emerging technology.

NOTES

^a MCQ International, Sydney, NSW 2000 Australia <http://www.mcqi.com.au/>.

^b Microsoft Corp., Redmond, WA 98052-6399 USA <http://www.microsoft.com>.

^c Blackboard, Inc., Washington, DC 20036 USA <http://www.webct.com>.

^d Apache Software Corp., Forest Hill, MD 21050-2747 USA <http://tomcat.apache.org/>.

^e Apache Lucene Project, Forest Hill, MD 21050-2747 USA <http://lucene.apache.org/>.

^f ImageMagick Studio LLC <http://www.imagemagick.org>.

^g Red Hat Inc., Raleigh, NC 27606 USA <http://www.redhat.com/>

^h Sun Microsystems, Inc., Santa Clara, CA 95054 USA <http://www.sun.com/>

	REFERENCES

TOP ABSTRACT PROJECT OVERVIEW PROJECT OBJECTIVES DATABASE FUNCTIONALITY SYSTEM ARCHITECTURE AND... USING OLIVER OLIVER SHOWCASE DISCUSSION REFERENCES

 

1. University of Bristol. UK: University of Bristol, 2004 Bristol BioMed Image Archive home page <http://www.brisbio.ac.uk/index.html> Accessed 06/04/07.
2. The Higher Education Academy [THEA]. Centre for Bioscience ImageBank home page <http://bio.ltsn.ac.uk/imagebank/>. THEA, n.d Accessed 06/04/07.
3. Gonzalez-Couto E, Hayes B, Danckaert A. The life sciences Global Image Database (GID). Nucleic Acids Res 29:336–339, 2001.[Abstract/Free Full Text]
4. Phillips R, Pospisil R, Richardson JL. The use of QTVR image database for teaching veterinary radiology and diagnostic ultrasound to distance education students. Aust J Educ Tech 17:94–114, 2001.
5. Siegel E, Reiner B. Electronic teaching files: seven-year experience using a commercial picture archiving and communication system. J Digit Imaging 14:(suppl. 1), 125–127, 2001.[Medline]
6. Blum JM, Aboulafia A. System design and implementation of a national image registry for orthopaedic oncology image management, research and teaching. Inform Syst Front 5:421–427, 2003.[CrossRef]
7. Lam AKY, Veitch J, Hays R. Resuscitating the teaching of anatomical pathology in undergraduate medical education: Web-based innovative clinicopathological cases. Pathology 37:360–363, 2005.[CrossRef][Medline]
8. Trumm C, Dugas M, Wirth S. Digital teaching file: concept, implementation, and experiences in a university setting. Radiologe 45:724–734, 2005.[CrossRef][Medline]
9. SNOMED International. SNOMED International, 2007 SNOMED home page <http://www.snomed.org/> Accessed 06/04/07.
10. National Library of Medicine [NLM]. Bethesda, MD: NLM, 2007 Medical Subject Headings (MeSH) <http://www.nlm.nih.gov/mesh/> Accessed 06/04/07.

This Article Abstract Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Search for Related Content PubMed PubMed Citation Articles by McGreevy P Articles by Miller N

AAVMC APPRECIATES THE SUPPORT OF OUR TWO PATRONS, HILL'S PET NUTRITION AND BAYER ANIMAL HEALTH, WHO IN COMBINATION ARE FULLY SUPPORTING THIS SITE.