Managing Knowledge in the Development of a Decision-Support System for the Investigation of Dementia
نویسنده
چکیده
There is a need to improve dementia care in Sweden. The main issues discussed are how to improve the competence of medical personnel and the quality of diagnosis and intervention. This thesis describes the development of a clinical decision-support system (CDSS) for the domain, which is considered one way to support diagnosis of cognitive diseases and choices of interventions, as well as to provide a continuing medical education and to disseminate clinical guidelines. The clinical practice of dementia investigation and cognitive processes in clinical decision making were investigated. A model of the clinical reasoning process, as performed by an expert physician in the domain, was implemented and evaluated. The domain experts found the process being well in accordance with how they work, while the primary care physicians were less familiar with the procedure. The domain knowledge was assessed and integrated into the model of reasoning. Since none of the clinical guidelines for the domain are suf cient for diagnosis on its own, sets of guidelines have been identi ed and formalised. As a result, a distinction has been made between typical and atypical cases of patients. Manifested symptoms in a typical patient comply with a certain set of clinical guidelines where criteria for one single diagnosis are met. These criteria are implemented into the support system using if-then rules. In atypical cases, the evidence is ambiguous or insuf cient for diagnosis according to this particular set of guidelines. The reasoning process proceeds in these cases with additional clinical guidelines and inferences which handle defeasible reasoning. A key issue is how the uncertainty expressed in the additional guidelines is best represented in the CDSS. Techniques developed for the purpose of handling ambiguous and incomplete knowledge have been reviewed in the thesis in order to nd an approach suitable for representing the characteristics of the domain knowledge. A larger emphasis has been put on techniques which allow the user to trace the inferences and which provide an overview of grounds for hypotheses framed in different contexts. The approach taken in the thesis is to let the different sets of guidelines represent different contexts, where features and sets of features are allowed different values, depending on context. An argumentation framework has been adapted and exempli ed in the formalisation. Further, the design and evaluations done of the CDSS are described in the thesis, motivated and analysed in terms of activity theory. The work will continue with the implementation of the decision-support mechanism for atypical cases and evaluations of the system when it is integrated into clinical practice.
منابع مشابه
A Model-Driven Decision Support System for Software Cost Estimation (Case Study: Projects in NASA60 Dataset)
Estimating the costs of software development is one of the most important activities in software project management. Inaccuracies in such estimates may cause irreparable loss. A low estimate of the cost of projects will result in failure on delivery on time and indicates the inefficiency of the software development team. On the other hand, high estimates of resources and costs for a project wil...
متن کاملSmart Diabetic Screening and Managing Software, A Novel Decision Support System
Background: Diabetes is a serious chronic disease, and its increasing prevalence is a global concern. If diabetes mellitus is left untreated, poor control of blood glucose may cause long-term complications. A big challenge encountered by clinicians is the clinical management of diabetes. Many IT-based interventions such ad CDSS have been made to improve the adherence to the standard care for ch...
متن کاملProposing an Appropriate Architecture for Decision Support Systems in the Field of Complex Chronic Care: Micro-Services Based Software Architecture in Kidney Transplant Care
Introduction: Development and successfully implementation of knowledge based clinical decision support system (KBCDSS) in kidney transplantation (KT) could support decision-making, reduce cost and improve quality of care. For practical use of these systems, however, many challenges have to be met. Besides to well-recognized challenges of design and implementation of information systems in heal...
متن کاملProposing an Appropriate Architecture for Decision Support Systems in the Field of Complex Chronic Care: Micro-Services Based Software Architecture in Kidney Transplant Care
Introduction: Development and successfully implementation of knowledge based clinical decision support system (KBCDSS) in kidney transplantation (KT) could support decision-making, reduce cost and improve quality of care. For practical use of these systems, however, many challenges have to be met. Besides to well-recognized challenges of design and implementation of information systems in heal...
متن کاملAnomaly Detection Using SVM as Classifier and Decision Tree for Optimizing Feature Vectors
Abstract- With the advancement and development of computer network technologies, the way for intruders has become smoother; therefore, to detect threats and attacks, the importance of intrusion detection systems (IDS) as one of the key elements of security is increasing. One of the challenges of intrusion detection systems is managing of the large amount of network traffic features. Removing un...
متن کاملDevelopment of decision support tool for municipal solid waste management system in Iran based on life cycle assessment approach
Background and Objective: Various aspects including the environmental burdens, social and economic consequences of the waste management(WM) scenarios must be considered to come up with a comprehensive WM plan. Life Cycle Assessment (LCA) approach is a systematic method to quantify the environmental burdens of each WM scenario. Materials and Methods: This study used an LCA approach to develop a...
متن کامل