Skip to Main content Skip to Navigation
Theses

Vers une nouvelle génération d'outils d'aide à la décision s'appliquant à la prévention des risques lors de la prescription des antibiotiques : combinaison des technologies Web sémantique et de l'aide multicritère à la décision

Abstract : Motivated by the well documented worldwide spread of adverse drug events that are associated to antibiotics usage, as well as the increased danger of antibiotic resistance (caused mainly by inappropriate prescribing and overuse), we propose a general architecture for recommendation systems adapted for this kind of context and we develop a specific system for antibiotic prescription (PARS). The type of context that our architecture covers is characterised by highly risky decisions or decisions with high stakes. Such a system cannot be based on machine learning, since there are no available training data sets or case bases. However, rules of good practice and expert knowledge are available, therefore our system should be able to model and implement them. The proposed solution is intended to be used by a decision maker who must adapt his/her decision both to each subject’s specific needs and characteristics, as well as to different types of evolution. Our approach is based on the combination of semantic technologies with MCDA (Multi-Criteria Decision Aids). The decision support process involves two steps. First, by taking into account the specific application domain, the approach evaluates the relevance of each alternative (action) in order to satisfy the needs of a given subject. The first level of the decision support model aims to select all the alternatives that have the potential to fulfill the subject’s needs. Subsequently, the second level consists of evaluating and sorting the selected alternatives in categories according to their adequacy to the characteristics of the subject. We propose an approach that exploits the knowledge schemes of semantic web technologies (ontologies) and that structures the recommendation rules into a suitable sorting method: the MR-Sort with Veto. By doing so, our solution is able to link and match heterogeneous knowledge sources expressed by experts. In collaboration with the EpiCURA Hospital Center, we have applied this approach in the medical domain and more specifically in the prescription of antibiotics. The system’s recommendations were compared with those expressed in the guidelines currently in use at EpiCURA. The results showed us that PARS allows for a better consideration of the sensitivity of the patients to the adverse effects of antibiotics. Moreover, by taking into account the additional characteristics of the patients, the model is able to adapt to contextual changes (such as new antibiotics, side effects and development of resistant micro-organisms).
Complete list of metadatas

Cited literature [100 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-01684761
Contributor : Abes Star :  Contact
Submitted on : Monday, January 15, 2018 - 5:23:16 PM
Last modification on : Friday, October 23, 2020 - 4:58:36 PM
Long-term archiving on: : Tuesday, May 8, 2018 - 1:06:36 AM

File

Ben_Souissi_Souhir2.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01684761, version 1

Collections

Citation

Souhir Ben Souissi. Vers une nouvelle génération d'outils d'aide à la décision s'appliquant à la prévention des risques lors de la prescription des antibiotiques : combinaison des technologies Web sémantique et de l'aide multicritère à la décision. Intelligence artificielle [cs.AI]. Université de Valenciennes et du Hainaut-Cambresis; Université de Mons, 2017. Français. ⟨NNT : 2017VALE0027⟩. ⟨tel-01684761⟩

Share

Metrics

Record views

382

Files downloads

1309