Intelligent readmission reduction system to improve quality of care


Reducing hospital readmission rates is among the key priorities and desirable objectives of any hospital or health center. The rate of unscheduled readmissions less than 30 days after discharge is a quality indicator of the efficiency of hospital care. Therefore, there are several initiatives to reduce readmissions, such as the Hospital Readmission Reduction Program (HRRP) established in 2010, with the goal of improving patient care through financial penalization of hospitals with high readmission rates in the US.

The objective of REINGRES is to extract clinical knowledge both through retrospective analysis of patient clinical data (where natural language processing techniques are also implemented to extract additional information) and through the development of Machine Learning (ML) models to predict and identify patterns to (i) feed a decision support system for efficient patient and resource management and (ii) ) detect patients at high risk of readmission allowing preventive actions to improve hospital discharge management.
Furthermore, the ML algorithms implemented in this project not only anticipate patients' readmission risk, but also, according to different values of patient attributes, give different recommendations in order to work towards a more personalized medicine.

Vicomtech is working on the development of readmission predictive models as well as decision support tools to prevent possible readmissions and improve hospital quality. In addition, NLP tools are being used to obtain more information from medical records (history) and improve the quality of the models.

Looking for support for your next project? Contact us, we are looking forward to helping you.


Parque Científico y Tecnológico de Gipuzkoa,
Paseo Mikeletegi 57,
20009 Donostia / San Sebastián (Spain)

+(34) 943 309 230

Edificio Ensanche,
Zabalgune Plaza 11,
48009 Bilbao (Spain)

close overlay