EUNIS97, Grenoble (France) 9-11 September 1997

Ref: 031101

Multimedia Technologies in Universities

 M. Heydthausen, J. Knop, K. Vehlhaber

Introduction

Multimedia as one of the key technologies of the ending century will influence and challenge our universities to a high degree.

New technologies in higher education will change teaching and learning: Computer based training, computer assisted learning and teaching will be no longer slogans but will become common like other techniques as videoconferencing, lecture on demand or self-instructing courseware. But we think that there will be no real revolution in teaching and learning techniques. More likely there will be a (sometimes slowly) evolution, because not only techniques have to change but also teachers and students have to adopt themselves to that changes.

But not only education is challenged by the new techniques. University research and development and even administration will be influenced by multimedia. A broad variety of information that is now traditionally handled will be performed by multimedia information systems.

 In this paper we will sketch some multimedia techniques and their application in academic teaching and learning on the one hand and their application on a special example in university medicine on the other hand. We will discuss the possible benefits and will not hide the difficulties that arise in the context of their implementation.

Multimedia in teaching and learning

 Besides few critical voices the majority of authors in scientific journals dealing with questions of multimedia in teaching and learning share the opinion that multimedia techniques can be successfully used in university education. The nowadays used terms ,,Computer Aided Teaching (CAT)" and ,,Computer Aided Learning (CAL)" point out that these techniques influence both teachers and learners.

In the following section we will deal with three main questions:
 1. What actions must be taken to establish multimedia techniques?
 2. How does the technical scenario look like to provide an adequate technical infrastructure for multimedia?
 3. What organizational infrastructure can support multimedia in teaching and learning?

The scenario which we have in mind is that of the Duesseldorf University. Thus, the following statements and sections can not be valid for every university in every country. Some might step ahead others may lagging behind.

Actions for establishing multimedia techniques in teaching and learning

Acquisition and evaluation of available CAT- and CAL-programs

There are some sources for relevant products: publishers, research associations, previously made self-developments, public-domain-programs. The commercial conditions are quite different and vary from free of charge to very expensive.

There is also no correlation between price and quality.

 However, deciding to use these products depends on the evaluation by the lecturers.

Design and programming of own CAT- and CAL-programs

 The design and programming of own CAT- and CAL-programs includes the  construction of lecture materials (e.g. projection of multimedia - worked up foils) as well as the construction of individual education- or simulation programs. For the production of CAL-products for video processing should be set up a well equipped multimedia lab. Apart from tools for recording and digitalising pictures and sound there should exist workstations for a digital rework of multimedia material. The equipment of the multimedia lab must include author systems for construction of CAT- or CAL-products.

Distribution of CAT- and CAL-programs

Through license restrictions some commercial available multimedia products could only be distributed by original CD-ROM. Thus, CD-ROM players will be necessary for distributing these products. The technical limitations are obvious. Frequently used products without license restrictions should be made available to users by an appropriate server because of the easier technical handling and higher access-speed. Uncleared is the problem how such products could be used by students at their homes.

Electronic lecture-halls

On the university campus should be set up electronic lecture-halls and electronic seminar-rooms available for multimedia-supported lectures. They offer the following services:
· Online-access to computers for calculations, simulations, graphical representations. The computer monitor can be projected by a beamer to a screen in the lecture-hall.
 · Online-access to central servers holding videos, CBT-software, ... .
 · The lecturer has an interactive white board at his disposal.
 · There is a video-conference-system available in order to have live-conversations with external round of talks transmitting it into the lecture-hall.

Video conferences

The technology of the video-conference can be used in university education for the realization of the so-called teleteaching. Several applications are possible.

· Transfer of lectures to another place within the same university.

 · Cooperation between universities by transfer of lectures.

 · Real-time video conference between two lectures on the same subject in different universities.
 

Evaluation and integration of new multimedia techniques

 Through the dramatic development of multimedia techniques there are always new technologies that could be adopted for teaching and learning in universities. Some current keywords are Computer Supported Cooperative Work (CSCW) or Electronic White board. Further on a digital video-library is a very important service of education in the university. Such a video-library supported by a powerful video server can be used to set up a service called ,,lecture-on-demand".

Technical infrastructure

An adequate technical infrastructure can be developed from the activities pointed out in the previous section.

Most important parts of the technical infrastructure are:
· A high speed network with a big bandwidth.
· Multimedia labs for the design of own CAT- and CAL-products.
· Video servers keeping lectures, parts of it or other video information for supporting university education.
· Electronic lecture-halls.

Organizational infrastructure

In addition to technical infrastructure an organizational infrastructure is needed for efficient support of multimedia techniques. Part of this infrastructure is any university institution dealing with media in a wide sense: library, computing center, media centers (if existent) and every potential user of media.

To keep the organizational structure easy to survey it is planned to have a permanent multimedia working group in Duesseldorf consisting of:
· representatives of the different faculties,
· library,
· computing center and
· administration.

This working group is responsible for:
· progression of the multimedia concept for Duesseldorf university,
· planning, implementation and integration of new multimedia techniques,
· survey of the technical realization,
· coordination of all tasks concerning multimedia aspects in university.

Multimedia in Medicine

Not only aspects of teaching and learning are challenged by new techniques. University research and even administration will be influenced by multimedia. Our next examples for multimedia techniques are taken from medicine. Medicine combines various aspects for multimedia support: research, patient administration and the wide field of hospital information and communication systems.

Health care has lagged for a long time behind other domains supporting work by efficient computerized techniques. One exception are systems for financial transactions in hospitals and for patient admission, as related topics. These were the first systems supporting data processing in hospitals. Information systems for patient-related medical information came up later. Information in these systems is mainly handled in a traditional -non multimedia- manner.

The next section will sketch some aspects of changing traditional patient records to multimedia records and the benefits connected with this change.

Patient records

A patient record for a given patient contains all information concerning anamnesis, findings, treatment, and more of this given patient. It forms the basis for the continuity of the patient`s treatment and influences hereby the quality of health care substantially.

Patient records, however, are often found to be handled on a traditional manner. That means: a lot of paper, more or less exact and readable or complete, possibly referring to further data in X-ray archives, ECG departments, labs, ... .

During the last years a lot of efforts have been pushed into computerizing patient records 1). The term EPR (electronic patient record) describes and summarizes these efforts. In Europe a pre standard has been defined by the European Committee for Standardization (CEN).

Patient records in principle are multimedial constructs in an almost natural sense: The information stored in and ordered by an patient record consists of written material, pictures, graphics. The associated information can enclose for example film sequences (heart catheterisation), curve diagrams (ECG, EEG). Voice information is stored rarely in a traditional patient record by lack of an adequate medium. But a lot of information during the patient`s examination is acoustic information. Additionally, an EPR is handled very interactively. With these details an EPR-system is qualified as a multimedia system.

For the user (doctors) of an multimedia EPR-system arise a lot of advantages:
· Reduction of written materials. Representation of information by means of a graphical patient record.
· Direct access even to time dependent media like X-ray films.
· Possibilities to integrate other media like videos or voice information to the patient record.

Visual Information Retrieval

For searching information in an EPR-system the user needs an effective tool. An information retrieval system is expected to help a user specify an expressive query to locate relevant information. The role of the emerging field of visual information retrieval (VIR)2) systems is to go far beyond text-based descriptors to store, and retrieve this ,,imagery-based" information content in visual media.

Visual information

There are two kinds of information associated with a visual object (image or video): information about the object, called its metadata, and secondly information contained within the object, called visual features. Metadata is alphanumeric and generally expressible as a schema of relational or object-oriented database. Visual features are derived through computational processes - typically image processing, computer vision, and computational geometric routines - executed on the visual object.

The simplest visual features that can be computed and retrieved are based on pixel values of raw data, and several early image database systems used pixels as the basis of their data models. A pixel-based model suffers from several drawbacks. One is, that variations in illumination and other imaging conditions affect pixel values drastically, leading to incorrect query results, but significant video segmentation results can be obtained by measuring pixel differences over time.

Most applications for VIR fall between automated pixel-oriented information models and fully human-assisted database schemes. They do not require pixel-level queries; nor are they constrained to only a few object classes. For this middle-of-the-spectrum applications, visual information can be defined in terms of image-processing transformations computed on the visual object. In most of these middle-of-the-spectrum applications even the system's designer needed training to perform effective retrieval. In medical-image databases, fully automatic feature extraction is still a research problem. The general experience is that completely automated image analysis works well only for small, controlled domains and it is very computation intensive.

Moving from images to videos in VIR adds several orders of complexity. Most research and commercial efforts take the following approach: Consider a video clip as a large number of image frames with progressively varying image content. Videos contain three kinds of motion information: one due to movement of the objects within the scene, second due to motion of the camera, and third due to special post-processing effects, like image warping. Ideally, a video information system integrates motion and frame information into a single computational framework, but current research is not there yet.

Summarizing we can say that many aspects of VIR systems are important for the application of multimedia techniques im medicine but they are not yet properly understood. Especially for the retrieval of multimedia databases there are still a lot of research problems.

Assessment of Cost/Benefit Relation

An estimation of a cost/benefit relation for a given technique has two faces: First the specification of costs and secondly an understandable description of what could be the benefits of this specific technique.

The question of the costs for the technical implementation is easy to solve. In contrast to this, the description of benefits is a hard job. What is the benefit of the multimedia support for teaching and learning? What amount (in ECU e.g.) is it worth? Being not able to answer these questions we will choose another way to describe benefit. We will try to answer the following question: what could be the obstacles, what are the difficulties for multimedia techniques to produce benefits?

Costs

The following table gives some details of necessary expenses (in ECU) building up the multimedia infrastructure for Düsseldorf university.
Estimated costs for:                           Amount (in ECU):  
                                                                 
Investments                                                      
        Electronic lecture-hall (per piece)             110.000  
        Multimedia-lab (per installation)                90.000  
        Video server                                    125.000  
                                                                 
Future costs for investments (5%, per year)               9.000  
                                                                 
Man power (per year)                                             
        Multimedia-lab                                  100.000  
        Video server                                     25.000
This table does not contain costs for network infrastructure which is a condition sine qua non for any multimedia technique and it does not contain costs for further extensions of the underlying network.

These few examples show that the costs are enormous: one electronic lecture-hall is not enough for a large university no more than the installation of one multimedia lab. Thus, the respective costs in the table have to multiplied by an adequate factor. Over the years the costs for investments still stay relatively small. The real financial problem are the costs for the supporting man-power.

Obstacles and difficulties

For efficient support of teaching and learning by multimedia techniques it is important to integrate these techniques as normal tools in preparing and offering lectures and as a normal way to achieve knowledge. A lot of possible obstacles must be overcome:

1. Old habits.
 2. Change in self-understandig: the professor is not a teacher but a learning facilitator.
 3. Preparing multimedia for teaching requires a lot of time.
 4. Support of teachers in preparing multimedia requires a lot of man-power and for that reason a lot of money.
 5. ...

Points 3 and 4 are only technical obstacles that could be solved by means of time, man-power and/or money. Real problems are old habits or necessary changes in self-understanding of teaching persons. The ,,real" integration of multimedia techniques means a fundamental change of teaching methods also.

For multimedia support in other areas of interest (like administration or medicine) we have the same technical and financial restrictions like in teaching and learning. Specific for these areas is -as pointed out already in the section about visual information retrieval- the need for further research. Often there is a great uncertainty how multimedia techniques could support a given area of interest.

References:

  1. David G. Kilman and David W. Fuslund, An International Collaboratory Based on, Virtual Patient Records, Communications of the ACM, August 1997/Vol. 40, No. 8
  1. Amarnath Gupta and Ramesh Jain, Visual Information Retrieval, Communications of the ACM, May 1997, Vol. 40, No. 5


Computer Center , Heinrich-Heine-University,
Universitaesstr. 1, D-40225, Düsseldorf, Germany
E-mail:  knop@uni-duesseldorf.de, heydth@uni-duesseldorf.de

Copyright EUNIS  1997  Y.E.