HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Master Thesis

Processus de conception et de validation d'un outil d'intelligence artificielle pour la conciliation médicamenteuse dans les essais cliniques

Abstract : In order to secure the therapeutic pathway of patients in clinical trials and to ensure the quality of the data collected, the hospital pharmacist performs a medication reconciliation (MR) for each new patient included in a clinical trial protocol that issues list of Prohibited Drugs or drugs to be Monitored (PDM), with a risk of interaction with the experimental drug. The realization of the medication review and the pharmaceutical analysis with the search for PDM must be carried out rapidly before the inclusion of patients and according to the information issued by the sponsors. Other issues concerning inter-pharmacist reproducibility and the multitude of sources used for pharmaceutical analysis have raised the need to create a digital tool capable of aggregating these multiple sources and harmonizing practices. Through a collaboration between pharmacists from a Cancer Center and pharmacists from a startup specialized in the field of artificial intelligence (AI) applied to health, the development and validation processes of an AI tool for MR in clinical trials were set up. The pharmacist responsible for the devlopment process wrote specifications for building data algorithms that would allow pharmacists to construct drug groups incorporating the PDM of each clinical trial, automatically detected during pharmaceutical analysis. It was also specified the creation of a quick search module for pharmaceutical advice requests, the constitution of a dashboard with different monitoring indicators (e.g.: number of conciliations, number of PDM) and other functionalities. In view of this project to generalize MR to clinical trials, two research protocols were developed to validate the automated PDM detection method and to evaluate the tool's performance in real life in several centers.
Document type :
Master Thesis
Complete list of metadata

Contributor : Odonto Ub - Connect in order to contact the contributor
Submitted on : Thursday, January 27, 2022 - 11:05:09 AM
Last modification on : Tuesday, February 1, 2022 - 3:39:42 AM


 Restricted access
To satisfy the distribution rights of the publisher, the document is embargoed until : 2023-06-02

Please log in to resquest access to the document


  • HAL Id : dumas-03545351, version 1



Domitille, Oddon. Processus de conception et de validation d'un outil d'intelligence artificielle pour la conciliation médicamenteuse dans les essais cliniques. Sciences du Vivant [q-bio]. 2021. ⟨dumas-03545351⟩



Record views