|
|
 |
 |
| |
 Printable version |
Drug-drug interactions detection systems improvement |
Journal de Pharmacie Clinique. Volume 28, Number 4, 213-20, octobre-novembre-décembre 2009, Article original
|
 Résumé
Texte intégral
|
Author(s) : F Mille, C Schwartz, S Roy, F Brion, P Degoulet, MC Jaulent |
Summary : Objective: The aim of this study is to propose changes detection systems drug-drug interactions in order to improve their clinical relevance. Methodology: The methodology that we followed, is threefold. First, we conducted an analysis of the functioning of a system. Secondly we have conducted a modeling study and acquisition of knowledge of IAM using the techniques of knowledge engineering and especially the techniques of documentary engineering. Finally, we have sought, through a descriptive study, if there is a link between alerts (combination), the service where is reported warning and response (acceptance or rejection) of the user. It appears that the connection is likely, paving the way for the adaptation of alerts according to the user. Results: We created a classification of overridden alerts. Thus this classification make it possible to creat an algorithm making it possible to decrease the sensitivity to background noise of drug-drug interaction detection systems. Knowledge engineering enabled us to build a DDI model. Techniques of documentary engineering made it possible to build a knowledge base on drug-drug interactions. This knowledge base is necessary in order to use the algorithm. Finally we find a link between alerts, the service and response of the user. Thus there is habits of prescription. Conclusion: The theoretical results obtained encourage further research in this area, to implement the algorithm and the knowledge base developed. They incite to extend the scope of research in the field of data-mining. |
Keywords : computerized decision support systems, drug-drug interaction, improvement |
|
