Muyal-Nez, una plataforma de construcción de sistemas de ciencias de datos médicos para procesos de toma de decisiones en el sector salud

Abstract

In recent years, data science systems and big data itself have turned into complex technologic assets that have started showing their possibilities to work with massive amounts of medical information, to support decision making processes such as prevention, detection, diagnosis, and prognosis, for instance, during cancer treatments. These groundbreaking technologies have triggered a revolution in scientific environments, but also in public health management. For example, a public health organization such as the Instituto Nacional de Rehabilitación (INR), in Mexico, has a collection of 45 million of medical images amounting to 43 Terabytes (TB) that require storage, transportation, and process, to help physicians during decision making processes. Nevertheless, the construction of data science system represents a huge challenge for this type of institutions as they do not only have to manage large volumes of data, or interconnect and develop complex artificial intelligence applications (such as machine learning or data mining applications), but also, they are compelled to accomplish a very strict set of regulations concerning security issues during the management of personal data. In this paper, we present the experiences and learnings obtained from the design, implementation of Muyal-Nez, which is a set of tools oriented to the design and deployment of data science infrastructure, that fits with the particular requirements of any public health organization on the aforementioned conditions.