Search

Artificial Intelligence algorithms’ application to medical data has gained relevance due to its powerful benefits among different research tasks. Nevertheless, medical data is heterogeneous and diverse, and these algorithms need technological support to tackle these data management challenges. The CARTIER-IA platform enables different roles (including principal researchers, IA developers and data collectors) to unify medical data, both structured data and DICOM images, and to apply Artificial Intelligence algorithms to them in a straightforward way through an online web application. However, given the diversity of roles involved in the platform, it is essential to account for its usability. It is necessary that users feel comfortable using the platform as relevant and complex tasks are carried out through its different services (such as the application of algorithms to the stored data, the manual edition of medical images or the visualization of structured data). This work presents a heuristic evaluation of the CARTIER-IA platform to improve its interaction mechanisms and get the most out of its functionalities.

A correct label classification of data allows improving healthcare processes and research. However, labeling is a difficult and expensive process, which limits its use and quality. We propose a proof of concept based on Natural Language Processing and active learning, in order to automatically structure information from a text in Spanish in the field of echocardiography.

Brugada syndrome is characterized by a genetic mutation that leads to a reduction in sodium inward current in phase 0 of the cardiac action potential. However, electrophysiological abnormalities that underlie the phenotypical presentation of this disease remain unclear. The objective of this study was to characterize the spatial electrophysiological changes produced by a reduction in sodium current. Epicardial transmembrane voltage optical maps of 10 Wistar rat hearts isolated and perfused in a Langendorff system were analyzed. For each heart, mean transmembrane action potential duration (APD50) and action potential up-stroke rise time (RT) were measured in the right ventricle (RV) right ventricular outflow tract (RVOT) and left ventricle (LV). Measurements were obtained at baseline and after injection of Flecainide 10uM, a potent sodium channel blocker. Flecainide injection produced a significant increase in APD duration and in the RT at each region with respect to basal conditions. However, the increase in the duration of the APD was higher in the RV compared to the LV (17.34 ± 6.67ms vs. 7.09 ± 6.39ms, p

The difficulty in obtaining research models of chronic atrial fibrillation (AF) is one of the main constraints to make progress in the elucidation of the mechanisms of perpetuation of this arrhythmia. The aim of this study is to present a new model of in vitro AF with different degrees of complexity similar to those observed during AF with varying degrees of structural remodeling. In the present study, optical calcium mapping is used to characterize the degree of disorganization of the reentrant activity in HL-1 confluent cell cultures. Specifically, the number of simultaneous rotors is linked with the time and degree of inhomogeneity of cell cultures. HL-1 cell cultures showed electrophysiological characteristics dependent on their degree of confluency resembling processes with varying degrees of complexity as it occurs during AF with varying degrees of structural remodeling. This model could be useful for studying the effect of remodeling on the fibrillatory process and to evaluate new antiarrhythmic drugs.