This paper aims to discuss the use of edge computing in medicine with a focus on the analysis of ECG biosignals. Edge computing is a novel paradigm which aims to perform computations (or at least most of them) near the data source achieving some advantages over classical centralized or distributed approaches (e.g. on cloud). After introducing edge computing and the novel NVIDIA Jetson device, the paper presents a use case regarding the classification of ECG biosignals on such a device. Several experiments were conducted to show main differences between traditional and edge-based data analysis approaches. Performance evaluation showed little differences between a traditional approach given the power constrained scenario of edge device. Main results of the paper include an overview of the edge computing paradigm, and a first performance evaluation of deep learning applications on a NVIDIA Jetson device.

Edge-based Deep Learning in Medicine: Classification of ECG signals

Barillaro L.;Agapito G.;Cannataro M.
2022-01-01

Abstract

This paper aims to discuss the use of edge computing in medicine with a focus on the analysis of ECG biosignals. Edge computing is a novel paradigm which aims to perform computations (or at least most of them) near the data source achieving some advantages over classical centralized or distributed approaches (e.g. on cloud). After introducing edge computing and the novel NVIDIA Jetson device, the paper presents a use case regarding the classification of ECG biosignals on such a device. Several experiments were conducted to show main differences between traditional and edge-based data analysis approaches. Performance evaluation showed little differences between a traditional approach given the power constrained scenario of edge device. Main results of the paper include an overview of the edge computing paradigm, and a first performance evaluation of deep learning applications on a NVIDIA Jetson device.
2022
Artificial Neural Networks
Deep learning
Edge computing
Medicine application
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12317/101268
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? ND
social impact