Periodic Reporting for period 1 - MEDAS (MedAS: a Machine learning enabled Clinical Decision Support System to prevent prescription errors and improve patient safety) Período documentado: 2017-05-01 hasta 2017-08-31 Resumen del contexto y de los objetivos generales del proyecto Medication errors are a major problem for patient’s safety, causing harm (2M lives globally) and economic burden (€4,5B-€21,8B, depending on the country). ICTs such as Clinical Decision Support Systems can help reduce them, but they lack accuracy because they are unable to identify risky situations that are not covered by their fixed alarm systems. MedAS incorporates data form different sources and uses big data analytics and machine learning algorithms to identify previously undetectable prescription errors. In order to bring MedAS to TRL9 and commercialize it in the European market, we aim to improve its accuracy and patient surveillance capacities, as well as to extend its ability to detect new types of patients and the adjustment of MedAS to the markets we aim to reach. We will launch MedAS in Q1 2020, first in those countries that have higher rates of EHR adoption (Germany, UK and Italy) and then with others that advanced in health technologies application, reaching a global market share of 6,14% by Q4 20204. Trabajo realizado desde el comienzo del proyecto hasta el final del período abarcado por el informe y los principales resultados hasta la fecha To ensure commercial feasibility, we have conducted an in-depth analysis of our market, including its dynamics, its environment and key players, our end users and customers, and our competitors. We have also conducted and FTO analysis and defined our value proposition. We have also designed a communication strategy. To analyse the financial risks of launching MedAS, we have observed different scenarios and ensure financial viability in all of them. To assess the technical feasibility, we have stablished the tasks needed to improve our accuracy, our ability to identify new patients and ensure the adaption to the markets we want to reach. Avances que van más allá del estado de la técnica e impacto potencial esperado (incluida la repercusión socioeconómica y las implicaciones sociales más amplias del proyecto hasta la fecha) MedAS incorporates machine learning techniques and big data analytics to detect previously undetectable prescription errors. It is not based on fixed rules as its competitors and this enables it to identify more risky situations and prevent more errors. It is optimized to reduce false positive, reducing alert fatigue and minimizing disturbances to the clinical practice and facilitating the adoption of the technology by the clinician. It can be customized to the demographics and characteristics of the populations of each health care organization to improve accuracy.