Informations pratiques sur cette ressource

Description de la ressource pédagogique

Description (résumé)

The randomized clinical trial has long been the only source for gathering evidence about the effectivity and safety of a new medicinal product to gain market access. In the changing world of healthcare, where all new information about patients is flowing to the healthcare provider, the patient and Internet, patient or product-related information can now be integrated and used for obtaining proof of clinical effectiveness and patient benefit. New technologies for analysis, machine and automated learning are being integrated as well in the healthcare world. Both structured and unstructured information and data coming from scientific literature, patient social media, smart devices or electronic health records, among others, are all relevant and knowledge about the patient outcome can be extracted with advanced analysis methods.  Automated learning techniques allow for building decision-support machines that can determine whether a patient should be dismissed from hospital, whether they are taking their drug on time when at home or if they should be contacted by their physician as their disease is worsening.  This shift from product delivery to patient outcomes in healthcare comes at a time when the analytical technology, hardware and data standardization (e.g. OMOP, IDMP, CDISC) is allowing for massively parallel analysis of data. Understanding the patient health situation by analyzing their data fingerprint requires however an understanding of personalized medicine, respect for data privacy and an understanding of how the algorithms work. The gap between research and clinical practice remains wide but is narrowing ? analytics, and the ability to make it usable by physicians, can further close that gap without breaking the system economically or practically.  

"Domaine(s)" et indice(s) Dewey

• Généralités (500)
• Techniques (sciences appliquées) (600)