Nouvelle approche d’adaptation de modèle et d’intégration des données pour les recherches de médecine de précision

Abstract

Massive adoption of high-throughput genetics technologies, deoxyribonucleic acid (DNA) chips, and high-speed sequencing has accelerated complex research in precision medicine through the use of high-volume considerable amount of heterogeneous medical data. With such data at their disposal, researchers and clinicians can develop strategies tailored to each individual to provide more precise preventive and therapeutic treatments by targeting subgroups of patients on specific diseases from a large amount of data. genomics, clinical, demographic, and other types of information related to patients' lifestyles. Next Generation Sequencing (NGS) techniques play a key role in research in precision medicine; however, the complexity, diversity, heterogeneity, volume of data and the inability of the data analysis software available to perform their function easily on this dataset represents a challenge for direct use by the researcher. This thesis, from the field of software engineering, proposes a solution track for researchers in precision medicine. It proposes a new approach to adapt the data model to the research new data requirements, unify the storage location and format of all research data accompanied by a precision medicine research cycle aimed at simplifying the phase of preparing medical precision medicine analyzes. This research attempts to address the main issues facing researchers in this field. To validate this proposition, the results of a precise medicine research data analysis experiment were carried out using an experimental prototype on a real case of a research which consists in developing a predictive model that would make it possible to identify the patients at risk of developing Chronic Kidney Disease (CKD) disease in patients with type 2 diabetes (T2D).

Date	5 Oct 2020
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Alain April (Supervisor) & Christian Desrosiers (Co-supervisor)