Title: Requirement specification of an ontology-based semantic recommender system for medical prescriptions and drug interaction detection

Authors: Ali Asghar Safaei; Sayyed Saeid Safaei

Addresses: Department of Medical Informatics, Faculty of Medical Sciences, Tarbiat Modares University, Jalal-Al-Ahmad Highway, Tehran, Iran ' Islamic Azad University, Science and Research Branch, Tehran, Iran

Abstract: The prescription of proper drugs is very important for disease treatment and patient health. Search and analyse the resources to find the drug interactions is a time-consuming and difficult task. It is possible to accelerate this process and improve the quality of drug prescriptions using the life sciences information sources available on the semantic web. In this paper, requirement specification of a ontology-based semantic recommender system for medical prescriptions and drug interaction detection has been presented. The main functions of the system and its requirements were initially extracted and described in detail by observing the existing system in the hospitals, interviewing medical specialists, and analysing questionnaires. Then, an appropriate ontology was designed and implemented for the system by using Protégé to detect drug interactions. A prototype was developed using Java to evaluate the functions of drug recommendations and drug interaction detection. The system performance was evaluated under different scenarios compared with similar approaches for drug prescription and interaction detection. The evaluation results demonstrated the enhanced performance of the proposed approach in both drug prescription and interaction detection functions. This indicates the benefits of proposed system as a computerised physician ordering entry.

Keywords: semantic recommender system; adverse drug events; ontology; drug interactions; computerised physician ordering entry.

DOI: 10.1504/IJEH.2019.108565

International Journal of Electronic Healthcare, 2019 Vol.11 No.1, pp.1 - 24

Accepted: 31 May 2019
Published online: 20 Jul 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article