Borsa di Dottorato [scadenza 15/9/2020]: Cross-comparison and development of algorithms to assess performance detecting atrial arrhythmias from clinical continuous ECGs

Sono aperte fino al 15 settembre le application per il Dottorato di Ricerca in Informatica dell’Univeristà degli Studi di Milano.

Ulteriori informazioni sul bando possono essere trovate all’indirizzo:
https://www.unimi.it/it/corsi/corsi-post-laurea/corsi-di-dottorato-phd/aa-2020/2021-dottorato-informatica

In particolare si pone l’attenzione sulla borsa finalizzata al progetto finanziata da Amps LLC: “Cross-comparison and development of algorithms to assess performance detecting atrial arrhythmias from clinical continuous ECGs” di cui un abstract è riportato sotto.

Roberto Sassi

Università degli Studi di Milano,
Dottorato di ricerca in Informatica

————————————————————–

Cross-comparison and development of algorithms to assess performance detecting atrial arrhythmias from clinical continuous ECGs

The accurate detection of atrial tachyarrhythmias (ATS) and more specifically of atrial fibrillation (AF) is a topic of relevant clinical interest, particularly with the increasing spread of new acquisition devices. Current state of the art is based on several algorithms which, in most cases, have been validated on old and relatively small data sets. Even the advent and promising diffusion of Artificial Intelligence based modelling approaches, is most often flawed by the usage of relatively small datasets, which for these methods is even more critical.

The first aim of this project will be to make a detailed technical inventory of most relevant approaches for ATA detection, and to re-assess the respective accuracy on a large retrospective cohort of clinically collected data. The diagnostic accuracy of all methods will be cross-compared and will be determined applying different metrics, primarily based on MIT duration sensitivity and positive predictive accuracy, separately applied for the different types of ATAs.

Based on the experience collected with the technical survey, it is expected that the candidate will conceive new ideas, and develop new algorithms, possibly combining individual features, when available. More importantly, it is expected an extensive usage of the P-wave signal, often not considered in prior art.

————————————————————–