In recent years, clinicians can perform more in-depth analyses thanks to current advances in next-generation sequencing (NGS) and biological data's rapid growth and availability. They can combine biological data with other clinical and patient-specific information, such as electronic health records (EHRs), habits, ancestry and environmental factors, allowing them to analyze and find relevant information beyond what can be obtained with conventional methods. Artificial Intelligence (AI) models are becoming increasingly prevalent in biomedical research and clinical practice. These models have shown promise in many fields, such as risk stratification and modelling, personalized screening, molecular diagnosis of diseases, prognosis and prediction of response to therapies.
In this sense, this Thesis proposes integrating and analyzing multiple biomedical data sources to provide more advanced analyses to clinical experts. This Thesis proposes to solve some of the challenges encountered in the biomedical data ecosystem. To this end, a clinical research support tool has been developed to capture and integrate biomedical data in different formats (such as patient-specific information, genetic sequencing data, wearable sensor data in Big data environments, biomedical images, etc.). It also includes various analytical functionalities by designing new Artificial Intelligence strategies in terms of Optimization techniques, Machine Learning, Deep Learning and Explainable Artificial Intelligence. This Thesis proposes to address real-world and academic problems in the context of the biomedical data ecosystem.