Improving Nursing Education with Intelligent Systems

Dr. Peter Kokol


Kokol, P. (February 2004) Improving Nursing Education with Intelligent Systems. Online Journal of Nursing Informatics (OJNI). Vol. 8, No. 1. [Online]. Available at


Health care is one of the fastest growing areas in terms of care, treatment and the exploitation of new technology in Slovenia . There is a great need for new approaches ensuring that the education and work of health care professionals will be built upon the state of the art in nursing. As a consequence the educational, governmental and “industrial” institutions from Slovenia, UK, France, Austria, Italy and Greece have determined to work on above problem and the European Union (EU) has agreed to support two projects (NICE, ODIN) under the Phare Tempus Framework. The aim of this paper is to present one of the approaches developed, namely educational intelligent systems.


Information did become the key reference of the health care organisations through the world, including Slovenia . The amount of health and medical knowledge is increasing exponentially while our capabilities to manually deal with them are limited, making it impossible to store, retrieve, and extract new knowledge without the use of the information technology (IT). The use of IT to support nursing can provide significant benefits, if used properly and appropriate education given. For this reason educational, governmental and health care institutions from Slovenia, UK, Italy, France, Austria and Greece formed a consortium to work on the above problem and the European Union (EU) agreed to support two projects (NICE – Nursing Informatics and Computer Aided Design, ODIN – On Demand Intranet for Nursing) under the Phare Tempus Framework. The aim of this paper is to present one of the approaches developed, namely educational intelligent systems.

Computers in Nursing Education

Saranto et al. (1997) argued that computers and information technology should be incorporated into all nursing curricula. However, nursing programs have varied opinions as to how this material would be incorporated, if at all, into the curriculum. Computer Managed Instruction (CMI) has been used in nursing education (Habjanic, Kokol, Zorman & Japelj 1998) since the late 1960s (Kohl 1995, Hebda 1998). Its accessibility and self paced format tmake it well suited for both students and practicing nurses ( Kohl, Hebda) who can learn at their own pace and time. In addition. CMI also supports continuing education and distant learning.

Early applications of CMI employed room sized mainframes at large institutions, but the proliferation of microprocessors in the 1990 expanded the depth and breadth of instructional computing (Cambre & Castner, 1993, Williamson 1994,). Recent studies (1993, Athappilly 1994, Haus 1996, Hebda 1998) show that students using CMI have better average examination scores, improved ability for critical thinking (Williamson 1994) and enhanced computer literacy, facilitated decision making skills and positively affected achievements (Belfry 1988). In spite of these advantages many students and faculty (Khoiny 1995) remain reluctant to utilise CMI, but Haus reports that the perception and attitude toward computer managed instruction changes positively after actual use of CMI software packages.

After the introduction of the Intranet many researchers report its successful use in nursing education. Cunningham and Plotkin (1999) report Intranet use of in nursing clinical practice use and Todd (1999) noted very positive experiences with the use of E-mail in undergraduate teaching.

Athappilly, Durban et al. (1994) list three benefits for using CMI and multimedia tools in the educational process:

  1. quality multimedia presentation reduces coss despite large initial investments, the reduction of participants and instructors time are significant;
  2. the effectiveness of the teaching and learning is improved because of greater motivation, retention, and mastery of learning;
  3. production is improved because of increased satisfaction and enjoyment of learning.

But the majority of the current teaching tools for nursing education are based on the so called concept of Drill and Practice (Conrick 1998), motivated by the research of Skinner (1953) The major advantage of this type of learning is the immediate feedback to a student. There is no waiting period for correction and therefore students do not practice their mistakes. But some researchers suggest that after the novelty effect of drill and practice wears off and the motivational power is lost. The wear effect can be overcome if the educational package is adaptive and can be individualised. This can be achieved with the use of artificial intelligence and automated learning (Russel & Norvig 1995 ) which in addition offers the possibility to analyze the mistakes and explain the problem to a student. Thereafter we decided to employ the concept of intelligent systems to improve the learning process in nursing education.

Intelligent Systems

A prominent researcher in the field of machine intelligence Randy Davis of MIT's AI Lab describes intelligent systems as "power tools for thinking," and draws an analogy with mechanical tools that increase our physical abilities (cranes to lift vast amounts, telescopes to see farther, etc.). Intelligent systems are power tools for heavy lifting in the information world; in Davis's words they "complement, extend, and amplify our ability to think and solve problems in a manner analogous to the way that mechanical tools complement, amplify, and extend our physical capabilities."

In general we can define three different kinds of intelligent systems:

On the surface these systems differ in the degree of authority granted for action in the real world, the degree and type of co-operation with people, the type of intelligent assistance given, the degree of naturalistic intelligence in communication and co-operation, and the richness with which the systems acquire information from the world. However, at a very high level they share certain characteristics:

Intelligent Systems in Nursing

Browsing trough the recent literature we found following applications of intelligent systems in nursing:

Intelligent system approaches

Various approaches on which we can build intelligent systems do exist. Some of the most known are:

The decision tree approach has one big advantage in comparison with other machine learning techniques: very simple and clear representation of the path to acquired decision. It is mostly for that reason that we decided to use decision trees for educational purposes.

Decision trees

The algorithm for learning a decision tree is trivial and the representation of accumulated knowledge can be easily understood (Quinlan 1998). Namely, the decision trees do not give us just the decision in a previously unseen case - they also give us the explanation of the decision, and that is essential in educational settings.

A decision tree is induced on a training set, which consists of training objects (instances). Each training object is completely described by a set of attributes and a class label (category, outcome). Classes are mutually exclusive, what means that the training object can belong to only one class . Attributes can be continuous (numeric) or discrete. Continuous attributes are not suitable for learning a tree, so they must be mapped into a discrete space. A decision tree co ntai ns nodes and edges (links). There are two types of nodes. Each internal node (non-terminal node) has a split, which tests the value of the chosen attribute for the training objects, that have come into this node and according to that splits the training set. Each internal node has at least two child nodes. External nodes, also called leaves or terminal nodes, are labelled with outcomes. Nodes (internal and external) are connected with edges. Edges are labelled with different outcomes of test, performed in the source node. Number of edges that come out of the node depends on the number of possible outcomes of the test.

Unlike some other approaches the representation of a decision tree can be easily understood by a human. All tests in internal nodes of a tree can be determined so the importance of attributes can be obtained from the decision tree. This that is the way to take advantage of the decision trees even without using them for their primary task in decision making. According to above, decision trees can support the nursing education process in four ways (Kokol 1999):

  1. to represent the knowledge and decision making as a simple two-dimensional hierarchical model;
  2. to outline important factors needed for successful decision making
  3. to enable a nurse to use the decision tree (in the paper form or as a computer program) to learn, support and test their own decision making in new situations and
  4. using their own databases to construct the decision tree (using automatic learning) for their own cases.

Intelligent systems in Nursing Education – Two Examples

Breast Feeding

Knowledge about breast feeding is very important for midwifes in the manner that they can teach mothers i. e. which factors (attributes) do influence the duration and success of breast feeding. To find that out a midwife can generate a decision tree from a breastfeeding database (In Slovenia such a database was created by the project INSIST), selecting various attributes, various diagnosing attributes, etc. A sample decision tree induced form the breast feeding database is shown bellow.

[] type of additional feeding by age of 2 months?
|_ _[none] BREASTFED
|_ _[bottle] mother is employed?
| |_ _[YES] number of meals?
| | |_ _[5..6.75] using dummy?
||||_ _[yes] head circumference?
|||||_ _[31..32.75] PARTIALY BREASTFED
|||||_ _[32.75..34.5] NOT BREASTFED
|||||_ _[34.5..36.25] PARTIALY BREASTFED
|||||_ _[36.25..38] NOT BREASTFED
||||_ _[no] weight by delivery?
||| |_ _[2650..3032.5] BREASTFED
||| |_ _[3032.5..3415] PARTIALLY BREASTFED
||| |_ _[3415..3797.5] BREASTFED
| | |_ _[6.75..8.5] weight by delivery?
||||_ _[2650..3032.5] BREASTFED
||||_ _[3032.5..3415] type of additional feeding by age of 4 months?
|||||_ _[bottle] NOT BREASTFED
|||||_ _[bootle or spoon] education of mother?
||||||_ _[10.25..12.5] NOT BREASTFED
||||||_ _[12.5..14.75] BREASTFED
|||||_ _[spoon] BREASTFED
||||_ _[3415..3797.5] NOT BREASTFED
||||_ _[3797.5..41] NOT BREASTFED
| | |_ _[8.5..10.25] NOT BREASTFED
| |_ _[no] NOT BREASTFED
| |_ _[student] BREASTFED
|_ _[bottle or spoon] BREASTFED
|_ _[spoon] NOT BREASTFED

The first thing that a midwife can learn from such a decision tree is which attributes mostly influence the output of the decision tree, which can be prediction, diagnosis, class ification and similar. In our case the output is the prognosis if the infant will be breastfeed in the 6 month based on the history of first two months of his life. The most influential attributes leading to the positive outcome are:

?type of additional feeding by age of two months?
?mother is employed?
?number of meals?
?using dummy?
?head circumference?
?weight by delivery?
?education of mother?

The next interesting thing is a decision process that can be derived from the tree itself. For example from the tree we can read:

If the type of additional feeding by age of two months?

is none

then the infant will be breastfed by 6 month

This is quit an interesting finding for a midwife, According to it she has to suggest mothers not to give any additives to infants till second month if the mother has the intention to breastfeed till over sixth months.

An example of a more complex decision process is

If the type of additional feeding by age of two months?

is bottle


mother is employed?

is yes

then the infant is not breastfed.


If the type of additional feeding by age of two months?

is bottle


mother is student?

is yes

then the infant is breastfed.

Above finding seems very logical, but sometimes such straightforward conclusions are not directly evident nor from experience nor from the database. Many other more complicated decisions can be learned from the above decision tree, also the midwife can select different outputs from the tree to find out additional knowledge, test her knowledge, or just experiment with her data.


Diagnosing the type of the incontinence is another example of the useful educational intelligent system based on decision trees. A sample decision tree is shown bellow.

[0] NeuropathyPreventingTransmissionOfReflex
|____[0] LessenCapacityOfBladder
| |____[0] SensoryCognitiveOrMobilityDeficits
| | |____[0] WeakPelvicMuscles
| | | |____[0] DegenerativeChangesInPelvicMuscles
| | | | |____[0] NeurologicalImpairment
| | | | | |____[0] Without
| | | | | |____[1] Reflex
| | | | |____[1] UrinaryUrgency
| | | | |____[0] Total
| | | | |____[1] Urgent
| | | |____[1] NeurologicalImpairment
| | | |____[0] InabilityToReachToiletInTime
| | | | |____[0] HighIntraabdominalPressure(Obesity_GravidUterus)
| | | | | |____[0] Reflex
| | | | | |____[1] Stress
| | | | |____[1] Stress
| | | |____[1] Age(years)
| | | |____[60 .. 80] Reflex
| | | |____[80 .. 100] Total
| | |____[1] HighIntraabdominalPressure(Obesity_GravidUterus)
| | |____[0] SpontaneousVoiding
| | | |____[0] NoAwarenessOfBladderFilling
| | | | |____[0] Without
| | | | |____[1] Functional
| | | |____[1] Functional
| | |____[1] Total
| |____[1] WeakPelvicMuscles
| |____[0] Urgent
| |____[1] Age(years)
| |____[60 .. 80] Urgent
| |____[80 .. 100] Without
|____[1] LessenCapacityOfBladder
|____[0] Alcohol
| |____[0] Total
| |____[1] Sex
| |____[m] Total
| |____[f] Functional
|____[1] Urgent

The decision process learned from the tree, regarding the diagnosing process in this case is

Absence of NeuropathyPreventingTransmissionOfReflex and

absence of LessenCapacityOfBladder and

absence of SensoyCognitiveOrMobilityDeficit and

absence of WeakPelvicMuscles and

absence of DegenerativeChangesInPelvicMuscles and

absence of NeurologicalImpairmen

indicates the definition Without incontinence.

Another important fact learned from the tree above is that instead of using 46 attributes defined in the nursing process by V. Henderson only 12 attributes are really enough for making almost 100% accurate diagnosing.


The aim of this paper was to show that with the use of information technology we can improve and better support nursing education processes. In the research presented we employed artificial intelligence, more specifically machine learning and decision trees. We showed that a nurse can improve here reasoning, find new facts or test her knowlede by generating and interpreting decision trees. It is our belief that in such way the use of computers in nursing ducation can become still more successful resulting in an overall health care process improvement.


Kokol, P. et al. (2001). Tempus ODIN report, Maribor: University of Maribor.

Kokol, P., Zorman ,M., Medos, T. (1999). Decision trees - a CMI tool in nursing education. Aust. Electron. j. Nurs. Educ., 4(2).

Kokol, P. et al, (1999). Nursing informatics education for the next millenium. FGCS, Future Gener. Comput. Syst., 15(2) 211-216.

Habjanic, A. et al,. (1995). CArE: a software package for computer-aided nurse education. Health Inform.J.j., 5(3) 119-123 .

Athappilly, K Durben, K. C., Woods, S. (1994) Multimedia Computing., Harrisburg: Reisman.

Ball, M. J. (1995). Nursing Informatics. New York: Springer.

Belfry, J. M. (1988), A Review of the effect of eomputer essisted instruction in nursing education, Computers in Nursing, March .

Cambre, M., . Castner, J. (1993). The status of interactive video in nursing education environments, Atlanta: FITNE.

Conrick, M. (1998). Computer based education: more than just a package. Australian Electronic Journal of Nursing Education 4:1 October.

Haus, K. (1996). Changes in perception of computer aided instruction, Research report, (http://parsons.ab.umd.cdu/~khaus/paper.html).

Hebda, T. (1998). A Profile of the use of CAI within baccalaureate nursing education, Computers in Nursing, November.

Khoiny, F. (1995). Factors that contribute to computer assisted instruction effectiveness, Computers in Nursing, July.

Kohl, J. E. (1995). Computer assisted instruction: Implications for achievement and critical thinking, Research report, (

Skinner, B. (1953). Science of Human Behaviour. New York: MacMillan.

Williamson, M. (1994). High tech training, Byte, December

Cunningham, H, Plotkin, K. (2000) Using the Internet in a Nursing Clinical Practicum Course: Benefits and Challenges, AEJNE Volume 5(2).

Saranto, K., Leino-Kilpi H., Isoaho, H. (1997). Learning environment in information technology: The views of students. Computers in Nursing 15(6): 324-332.

Russel, S., Norwig, P. (1995). Artificial Intelligence: A Modern Approach, New York: Prentice Hall.

Todd, N. (1998). Using E-mail in an undergraduate nursing course to increase critical thinking skill. Computers in Nursing 16(2): 115-118.

Mantas, J., Hasman A.: Textbook in Health Informatics – A nursing perspective, Amsterdam: IOS Press, 2002.

Quinlan, J. R. (1998). MiniBoosting decision trees, In Proceedings of Fifteenth National Conference on Artificial Intelligence, AAAI Press.

Author Bio

Dr. Peter Kokol

Professor and Deputy Dean, University of Maribor, FERI, Laboratory for System Design
Smetanova 17, Maribor, Slovenia.

Professor Peter Kokol obtained his bachelors degree in Engineering from University of Maribor and his Masters and Doctorate in Computer Science also from the University of Maribor. He is the Professor of computer science, the head of Research Institute and the dean for research at the University College of Nursing Studies. Since 2002 he is als the the Director of the independent Centre for Interdisciplinary and Multidisciplinary Studies and Research. He has written over 300 technical and research papers published in recognised international journals and major conferences and co-authored some textbooks. His main research interests are intelligent systems, complex systems, system and chaos theory, software quality and metrics, and medical and nursing informatics. He has acted as principal investigator on numerous international and national research projects. He is a member of ACM, IEEE and ASIS and some IMIA technical committees. He is the President of the IEEE Committee on Computational Medicine.