Relation Extraction (RE) is a crucial task which aims to identify and classify relations between entities in a given text, particularly, in the case of medical data, where understanding the relations between entities is essential for knowledge extraction. In the literature, Deep Learning (DL) models like BERT have shown promising results in RE tasks. Nevertheless, most of the proposed approaches do not incorporate pretraining before conducting finetuning on a new task, which could considerably improve performance. In this paper we present a novel deep learning-based approach for extracting semantic relations from unstructured medical datasets. Our approach takes place in three main phases: (1) a pretraining phase using unsupervised relation extraction techniques, (2) a fine-tuning phase employing supervised relation extraction, and (3) an inference phase for semantic relations’ deduction. Our relation extraction method is based on combining both unsupervised RE (during the pretraining phase) and supervised RE (during the fine-tuning phase). Our three phases-based approach, includes the hybridization of different BERT variants. The evaluation was carried out on the SemEval 2010 dataset. Experimental results show improvements in the evaluation metrics, which confirms the great interest of our proposal.

Medical question-answering systems have emerged as innovative tools for healthcare professionals and knowledge seekers. In this work, we leverage advanced Natural Language Processing (NLP) techniques to enhance the capabilities of such systems. Our approach involves the integration of an ontology layer to provide formal structures for medical data, resulting in improved answers through semantic concepts. Additionally, we harness the power of a vector database, reducing system training and response times and incorporate specialized medical deep learning model (BIOBERT) further which elevates the system’s performance. Our work contributes to the domain of medical question-answering by showcasing the potential of advanced NLP techniques and structured medical knowledge. With the usage of the following components in our proposed system: question analysis, answer creation, answer formulation, an ontology layer, a vector database and a session database, this work not only improves healthcare knowledge access but also enhances medical decision-making processes. Our contributions have led to remarkable improvements in accuracy, precision, recall, F score, and mean reciprocal rank (MRR).

Drug traceability is currently a very challenging area given the complexity of several issues, including drug quality and counterfeit medications. The counterfeited drugs have a major impact on human life, treatment outcomes and economic burden. To deal with these issues, we propose a semantic blockchain-based system for drug traceability that aims at detecting counterfeit drugs in order to improve the patients’ safety and quality of life as well as eliminating manufacturers’ potential loss and increasing their revenue. Our proposal is based on blockchain and semantic web technologies to enhance the representation capability of data in the pharmaceutical supply chain.