=

Interpretable Models

innovate in visual analytics to interpretate models, data and behaviors for the monitoring and spatiotemporal analysis of outbreaks and infections, including advances in visual control chart to generate early warnings of the progress of infections.

Hibrid Machine Learning Techniques

new hybridization techniques of machine learning algorithms for the creation of interpretable models of clinical prediction rules that identify subpopulations with risk of re-admissions and of appearance of MMR.
To create interpretable models, we will work on feature engineering and techniques such as classification by means of interpretable fuzzy rules.

Time Series

prediction of infections and antibiotic consumption, including visual tools for hypothetical forecasting.
In addition, considering the spatial dimension, we will adapt and extend spatial analysis techniques of graphs and networks to investigate thetemporal and spatial propagation of infections

Deep Knowledge

new multilevel pattern discovery techniques to generate clinical study hypotheses (deepknowledge).
Phenotype / genotype patterns will be obtained from multimodal data from the Electronic Healtcare Record and external bioinformatic and geographic sources.

Research Team

Multidisciplinary team of researchers from the University of Murcia, Spanish National Center of Epidemiology (CNE), Maribor University, AGH Univ. of Sc. and Technology and the University of Verona.

1

Years (1-2-3)

3

Publications

0

Patents/Copyright

34

Visits

Our Recent Milestones

These are our recent achievements from the academic, scientific and social point of view.

Pre-Kick of Meeting @ AIME 2019

During AIME 2019 international and advisory board members from Univ. of Maribor, AGH Sci ant Tech. Univ. , Zhenxiang Univ. and Univ. of Verona met and discussed the baby steps of the project.

10th Sept

Tools for optimising treatment

Outreach activity: presentation of research results during the Interhospital Sessions at Reina Sofia Hospital, Murcia.

11th Nov