A decision-theoretic network approach to treatment management and prognosis
Introduction
Medical patient management is a complicated process, usually involving a large amount of, possibly uncertain, information. It is, therefore, not surprising that many computer-based systems, including expert systems, have been developed during the past two decades to offer assistance in diagnosis, treatment management and assessment of prognosis. The overall goal of the development of these systems was improvement of patient care. However, despite considerable research efforts, few of these systems have actually reached the stage of clinical use. This lack of success in the medical field has been attributed to a number of different factors, both technical and social [1]. This paper focuses on a factor that belongs to the technical realm of expert systems: potentials of reuse of medical knowledge.
Recent model-based approaches to the development of medical expert systems offer substantial advantages in terms of flexibility and potentials of knowledge reuse, mainly due to the declarative nature of the knowledge embodied in such systems. Frequently employed formalisms to represent knowledge in model-based systems include logic and probability theory, among others. In medicine, using probability theory, possibly augmented with decision theory, is attractive, because it allows for explicitly representing the uncertainties and preferences that go with much of the knowledge used in medical decision-making. In particular, adopting probabilistic networks as a framework for building medical expert systems is appealing, because in addition to the advantages linked to using probability theory, qualitative relationships among variables can be represented, in the form of a directed acyclic graph. This feature not only simplifies probability assessment, but also promotes easy understanding of the resulting system. We shall refer to a system that incorporates a probabilistic network, possibly augmented with decision theory, as a decision-theoretic expert system. A knowledge base of such a system will be called a decision-theoretic network.
There are presently no comprehensive knowledge-engineering methodologies available that can be used in building a decision-theoretic network. In this paper, a further elaboration of a preliminary methodology for building decision-theoretic expert systems for treatment management, as described in [2], is discussed. It is also investigated how decision-theoretic expert systems can be used to deal with a number of questions clinicians are likely to find important: optimal treatment selection, prognostic assessment and generating profiles for specific treatment outcomes. Since all of these questions can be answered, in principle, using the same knowledge base of a decision-theoretic expert system, a practical example of knowledge reuse is arrived at.
As an example, medical treatment of patients with non-Hodgkin lymphoma of the stomach (gastric NHL) is discussed. Primary gastric NHL is a relatively rare malignant disorder, accounting for about 5% of gastric tumours. Until recently, the aetiology of gastric NHL was unknown; it is now generally believed that the main factor in the pathogenesis of this disease is a chronic infection with the bacterium Helicobacter pylori 3, 4. Therapy selection for gastric NHL is a complicated process, because only part of the patient findings necessary for therapy selection may be known at a particular stage of the disease, and knowledge of adverse reactions to particular treatments in patient groups may influence treatment selection significantly.
The structure of this paper is as follows. First, we discuss the process of building a decision-theoretic network, which is illustrated by the construction of a decision-theoretic network for gastric NHL. Next, various possible clinical applications of the resulting decision-theoretic expert system are considered. Finally, decision-theoretic networks are compared with conventional prognostic models in medicine.
Section snippets
Development of the gastric NHL decision model
In this section, first decision-theoretic networks are introduced; next, the successive steps in the design of the gastric NHL model are reviewed.
Reusing medical knowledge
The main goal of the development of the decision-theoretic model (influence diagram with associated probabilistic network) was to obtain a knowledge base for an expert systems that could be used to explore many different clinical questions. Although the resulting influence diagram is specialised for selecting optimal treatment for a given patient, which is highly dependent on a patient's preferences as expressed by a utility function, the associated probabilistic network offers much
Conclusions
A number of prognostic models of NHL, and of gastric NHL in particular, have been developed in the past. Examples of such models are those developed by Valicenti et al. [18], by Radaszkiewicz et al. [13], and by Azab and colleagues [12]. These models are based on univariate and multivariate analysis of data from patients with gastric NHL. Although these statistical techniques are useful for identifying relevant prognostic factors, the resulting models are of limited value from a clinician's
References (19)
- et al.
Gastrointestinal malignant lymphomas of the mucosa-associated lymphoid tissue: factors relevant for prognosis
Gastroenterology
(1992) Fundamental concepts of qualitative probabilistic networks
Artificial Intelligence
(1990)- E.H. Shortliffe, Knowledge-based systems in medicine, in: K.-P. Adlassnig, G. Grabner, S. Bengtsson, R. Hansen (Eds.),...
Knowledge acquisition for decision-theoretic expert systems
AISB Quaterly
(1996)- et al.
Helicobacter pylori gastritis and primary gastric non-Hodgkin's lymphoma
J. Clin. Pathol.
(1994) - et al.
Regression of primary low-grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobater pylori
Lancet
(1991) - J. Pearl, Probabilistic Reasoning in Intelligent Systems, Morgan Kaufman, San Mateo, CA,...
- et al.
Local computations with probabilities on graphical structures and their application to expert systems
J. Roy. Statistical Soc. (Ser. B)
(1987) Evaluating influence diagrams
Operation Research
(1986)
Cited by (11)
Human-environment interaction in the Baltic Sea
2014, Marine PolicyCitation Excerpt :Bayesian belief networks allow the joint use of objective and subjective information and, where hard data is absent, can perform calculations based on the opinion of specialists [3]. First used in the health service [5,6] and in artificial intelligence [7], Bayesian belief networks have more recently been used for aquatic environmental management, initially investigating the management of fisheries [8,9], and developing into studies of regeneration of forestry [10] and eutrophication [11]. Watershed management has also been investigated at river catchment scales [12,13], although in the present study Bayesian belief networks are used on a larger scale to investigate the influence of human activities on the marine environment of a regional sea.
A Bayesian network model for the diagnosis of the caring procedure for wheelchair users with spinal injury
2009, Computer Methods and Programs in BiomedicineExplaining clinical decisions by extracting regularity patterns
2008, Decision Support SystemsComputer-assisted decision support for the diagnosis and treatment of infectious diseases in intensive care units
2005, Lancet Infectious DiseasesCitation Excerpt :For example, the choice of antibiotics for VAP can be categorised so that relatively narrow-spectrum agents are chosen first, instead of broad-spectrum antibiotics.23 Bayesian networks can be also used for simulations, to determine the characteristics of patient groups and to investigate what will happen if a (potentially inappropriate) drug is prescribed.24 Probably the best-known medical decision support system for the treatment of nosocomial infections is the Health Evaluation by Logical Processing (HELP) system developed in the early 1970s in the LDS hospital in Salt Lake City, Utah, USA, and continuously improved since then.25
A probabilistic and decision-theoretic approach to the management of infectious disease at the ICU
2000, Artificial Intelligence in MedicineUSING BAYESIAN METHODS IN THE TASK OF MODELING THE PATIENTS' PHARMACORESISTANCE DEVELOPMENT
2022, Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Srodowiska