Problems in the Design of Knowledge Bases for Medical Consultation (original) (raw)
Related papers
Categorical and probabilistic reasoning in medical diagnosis
Readings in Uncertain Reasoning, 1990
Medical decision making can be viewed along a spectrum, with categorical (or deterministic) reasoning at one extreme and probabilistic (or evidential) reasoning at the other. In this paper we examine the flowchart as the prototype of categorical reasoning and decision analysis as the prototype of probabilistic reasoning. Within this context we compare PIP, INTERNIST, CASNET, and M YCIN-four of the present programs which apply the techniques of artificial intelligence to medicine. Although these systems can exhibit impressive expert-like behavior, we believe that none of them is yet capable of truly expert reasoning. We suggest that a program which can demonstrate expertise in the area of medical consultation will have to use a judicious combination of categorical and probabilistic reasoning-the former to establish a sufficiently narrow context and the latter to make comparisons among hypotheses and eventually to recommend therapy.
Diagnostic Reasoning Guided by a Decision-Support System: a Case Study
Proceedings of the European Conference on Cognitive Ergonomics 2017, 2017
A clinical decision-support system for dementia investigation was used in clinical practice. User information was collected based on interactions with the application. The aim of this study is to identify features in logged data that can be used for detecting learning and reasoning patterns in the user. A case of a physician who is novice to both the application and the dementia domain was studied and compared to the case of an expert physician using the system. Differences between them were found, and a clear pattern that indicates that learning takes place, both how to use the system and about dementia, was observed in the novice user. Further studies need to be conducted, focusing on whether patterns become stable over time, and with complementary methods that can detect reasons for observed behaviors. Software for automatic detection will be developed based on the results of this study.
A Framework for Medical Diagnosis using Hybrid Reasoning
Proceedings of the International MultiConference of …, 2010
Abstract—The traditional method of reasoning was rule-based reasoning (RBR). It does not use past experiences to reason. Case-based reasoning (CBR), on the other hand uses past experiences to derive results for new cases. Both rule-based reasoning and case-based ...
A Universal Model of Diagnostic Reasoning
Academic Medicine, 2009
Clinical judgment is a critical aspect of physician performance in medicine. It is essential in the formulation of a diagnosis and key to the effective and safe management of patients. Yet, the overall diagnostic error rate remains unacceptably high. In more than four decades of research, a variety of approaches have been taken, but a consensus approach toward diagnostic decision making has not emerged.
Medical Diagnosis and Expert Systems: Solving Problems, Problem Structures and Bayesian Networks
Medical Diagnosis and Expert Systems have a particular relationship. Since the birth of the Heuristic program from the Heuristic group headed by Feigenbaum and the beginning of the twentieth one century there were a lot of changes. In the first part of this paper I will introduce the topic of Medical Diagnosis and Expert Systems and how they are related. In the second part I will describe a small part of this story and how those changes can give us a different perspective about the development of Artificial Intelligence programs. The third part will be dedicated to Problem Solving and the structure of how we conceive problems. This part has its importance for explaining the differences of the solving methods for particular problem structures. The fourth part of this paper will characterize the Bayesian Networks and their links in structuring problems.
1981
RECONSIDER is an interactive computer program which produces a differential diagnosis given a list of patient attributes. The program's principal knowledge base is a corpus of 3,262 disease definitions represented in the form of structured natural language text. As these definitions were originally prepared for human use, RECONSIDER uses medical knowledge that is semantically identical to an information source that might be used by a physician. Thus, RECONSIDER can explain the inclusion of a particular disease in a differential by displaying the way in which the disease's definition relates to the list of patient attributes, and by ranking the strength of this relation relative to the rest of the differential. Use of RECONSIDER is illustrated on cases from the literature, and a case of pyruvate kinase deficiency (PK disease). Included for comparison are cases diagnosed by INTERNIST[1] and PIP[2], two well known diagnosis programs.
Medical Decision Aid: Logic Bases of the System SPHINX
The Advanced Computer Applications (ACA) project builds on IIASA's traditional strength in the methodological foundations of operations research and applied systems analysis, and its rich experience in numerous application areas including the environment, technology and risk. The ACA group draws on this infrastructure and combines it with elements of A1 and advanced information and computer technology t o create expert systems that have practical applications. By emphasizing a directly understandable problem representation, based on symbolic simulation and dynamic color graphics, and the user interface as a key element of interactive decision support systems, models of complex processes are made understandable and available t o non-technical users. Several completely externally-funded research and development projects in the field of model-based decision support and applied Artificial Intelligence (AI) are currently under way, e.g., Ezpert Systems for Integrated Development: A Case Study of Shanzi Province, The People's Republic of China. This paper gives an overview of some of the expert systems that have been considered, compared or assessed during the course of our research, and a brief introduction to some of our related in-house research topics. Kurt Fedra Project Leader Advanced Computer Applications 6.2.3.4 information Retrieval 6.2.3.5 Updating the Knowledge Base 6.2.3.6 Achievements 6.3 Automatic Learning-A Claim for the Future 6.3.1 Background and State of the Art 6.3.2 Objectives and Approach 7. Conclusion 8. References and Selected Bibliography