SMILE: smart monitoring intelligent learning engine: an ontology-based context-aware system for supporting patients subjected to severe emergencies. (2016)

• Record Type: Journal Article
• Authors: Khreiss, Malak; Zaarour, Iyad; Mcheick, Hamid
• Is Part Of: International journal of healthcare technology and management. Volume 15:Number 3(2016)
• Page Start: 194
• Page End: 209
• Publication Date: 2016
• Subjects: probabilistic reasoning -- uncertainty -- middleware -- intelligent learning -- pervasive computing -- remote healthcare -- ontology design -- Bayes reasoning -- Bayesian networks -- smart monitoring -- context awareness -- patient support -- severe emergencies -- heart attacks -- patient monitoring -- remote monitoring -- home care -- emergency handling -- healthcare emergencies -- emergency response -- emergency prevention -- lifestyles -- habits -- probabilistic modelling -- knowledge sharing -- knoweldge reuse -- decision making -- influence diagrams -- analytical hierarchy process -- AHP -- semantic logic; Medical technology -- Periodicals; Medical technology -- Management -- Periodicals; 610.28
• Languages: English
• ISSNs: 1368-2156

Abstract

Remote healthcare has made a revolution in the healthcare domain. However, an important problem this field is facing is supporting patients who are subjected to severe emergencies (as heart attacks) to be both monitored and protected while being at home. In this paper, we present a conceptual framework with the main objectives of: 1) emergency handling through monitoring patients, detecting emergencies and insuring fast emergency responses; 2) preventing an emergency from happening in the first place through protecting patients by organising their lifestyles and habits. To achieve these objectives, we propose a layered middleware. Our context model combines two modelling methods: probabilistic modelling to capture uncertain information and ontology to ease knowledge sharing and reuse. In addition, our system uses a two-level reasoning approach (ontology-based reasoning and Bayesian-based reasoning) to manage both certain and uncertain contextual parameters in an adaptive manner. Bayesian network is learned from ontology. Moreover, to ensure a more sophisticated decision-making for service presentation, influence diagram and analytic hierarchy process are used along with regular probabilistic rules (confidence level) and basic semantic logic rules.