During 2015 my research was developed manly in the framework of the below described lines, according with the proposal previously presented and including some new problems.

In terms of development and validation of tools for biomedical signal processing the acquisition of new MATLB toolboxes by CMUP allowed to use new tools both for modeling and analysis. A PEEC involving 2 master students entitled ““Modelação em MATLAB de series de variabilidade cardiovascular” took place for 6 months allowed to adapt the long memory and heteroscedasticity methodologies using ARFIMA-GARCH models using the new resources, in collaboration with other CMUP members. This new tools will be in future incorporated in the BioSigBrowser.

Additionally I have also been co-adviser of the grant “Bolsa de Investigação – mestre Development and implementation of a strategy for noninvasive Fetal Electrocardiograms quality assessment”, by MSc Carla Marisa Martins Oliveira, June to August 2015, in which a methodology for the assessment of Noninvasive Fetal ECG noise level was developed and preliminary validation has been performed.

The research on monitoring and data analysis in intensive care included the study of the statistical equivalence between data at different sampling rates, namely acquired with high resolution Holter recordings and typical bedside monitoring systems. This study have been done in the framework of the ongoing PhD of António Paulo Sousa (PDMA) “Auxiliary Prognostic Indexes in Critical Illness” and the preliminary results were presented at the The International Conference on Computing in Cardiology 2015, and published in the respective proceedings.

The collection of data from patients admitted in the pediatric intensive care unit of Hospital S. João (Porto) proceeded and currently the database counts with more than 70 patients and 150 recordings preprocessed and currently under analysis.

A new line of collaboration with Neurocritical intensive care unit of Centro Hospital S. João (Porto) was established.  Plateau waves (PW), defined by a sudden substantial elevation in intracranial pressure (ICP) while arterial blood pressure remains stable or varies modestly are common cerebrovascular phenomena in several neurological and neurosurgical conditions. These acute elevations of ICP may lead to dramatic decreases of cerebral perfusion pressure, reduction of brain blood flow and tissue oxygenation, contributing to secondary brain lesion. As the PW are probably a reflex of autonomic dysfunction, heart rate variability was used to study and characterized ICP plateau waves. The preliminary results were reported at the 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC’15), taking place 25th-29th  August 2015 , in Milan, Italy, and published in the Conference Proceedings.

The results of the multivariate model based in Independent Component Analysis (ICA) for estimate the bathymetric value Douro River estuary (Porto, Portugal) from an IKONOS-2 image developed and validated was presented at the Conference on Remote Sensing for Agriculture, Ecosystems, and Hydrology, part of SPIE Remote Sensing 2014 and published in the proceedings [TAG14].

The multivariate model based in Independent Component Analysis (ICA) for estimate the bathymetric value Douro River estuary (Porto, Portugal) from an IKONOS-2 image developed and validated during 2014 for  was further enhanced during an ungoing PEEC (Programa de Estágios Extra-Curricular) entitled "Estimativa da batimetria do estuário do rio Douro usando imagens de satélite e métodos estatísticos (componentes principais e componentes independentes)" and preliminar results presented at IJUP 2015 ("8º Encontro de Investigação Jovem da Universidade do Porto"). A full paper in this topic is under preparation.