Conference Paper

A Fuzzy Approach to Temporal Model-Based Diagnosis for Intensive Care Units.

Conference: Proceedings of the 16th Eureopean Conference on Artificial Intelligence, ECAI'2004, including Prestigious Applicants of Intelligent Systems, PAIS 2004, Valencia, Spain, August 22-27, 2004
Source: DBLP
Download full-text


Available from: Jose T. Palma
  • Source
    • "However, we decided to keep the expressivity of the TBM. Thus, in our model, this improvement in efficiency is achieved by the reduction of the search space through the use of the subsumption process [10]. This process is based on the consideration that it is possible that the cause that produces a given finding already exists in the diagnosis solution, thus explaining any other finding. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In recent years, medical knowledge has grown importantly, achieving high levels of specialization. The underlying models of the domain knowledge have become more complex, even more so when the temporal dimension is considered. This scenario requires the de-scription of behavioural models and developments of Knowledge-Based Systems (KBS) to deal with these new requisites. This paper describes our experience in designing and developing a Model-Based System for decision support in Intensive Care Units (ICUs). A set of tools are presented in order to facilitate the knowedge acquisition and diagnosis issues. Difficulties are also described in applying model-based paradigm in medical domains.
    Full-text · Article ·
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this project (MEDICI: Inpatient/Outpatient Monitoring for Diagnosis and Medical Research in Ischaemic Cardiopathy: Tools for Acquisition, Visualization, Integra- tion and Knowledge Discovery) is the design, development and implementation of a set of tools for inpatient and outpatient monitoring of patient sufiering Ischaemic Cardiopathy. These tools will provide support for both medical care and clinical research tasks. The work is organized around three development lines: a) Subsystems for data collection in difierent scenarios, including clinical history data, monitored electric and hemodynamic signals and data related to remotely monitored signals outside the hospital. b) Techni- ques for integrating the collected information and presenting it as a unique virtual clinical history associated to the patient; and c) tools for the extraction and validation of medical knowledge.
    Full-text · Article ·
  • [Show abstract] [Hide abstract]
    ABSTRACT: Over the last decades, Artificial Intelligence has approached the decision support system design in medical domains by capturing the knowledge and configuring it in knowledge intensive software systems. Model-based diagnosis is one of the techniques which has produced the best results, such as diagnosis intelligent systems in the realm of medicine. In this domain, one of the key factors is the temporal dimension. This variable enormously complicates the design of such systems, and in particular the process of getting a reliable diagnosis solution. This paper presents a Diagnosis Abductive Algorithm based on Fuzzy Temporal Abnormal Model. This algorithm provides a solution for the above problem by the description of its dianosis explanation, allowing an approach based on the Possibility Theory for the evaluation of the diagnosis hypotheses. KeywordsModel-based Diagnosis-Temporal Reasoning-Possibility Theory
    No preview · Conference Paper · Jan 2005
Show more