Neural Network Detection of Atrial Fibrillation by Lorenz Plot Images of Interbeat Interval Variation

Masaya Kisohara, Yuto Masuda, Emi Yuda, Junichiro Hayano

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

We developed artificial intelligence (AI) atrial fibrillation (AF) detection system using the big data of Allostatic State Mapping by Ambulatory electrocardiogram (ECG) Repository (ALLSTAR). Detection of atrial fibrillation (AF) is important for preventing acute cerebral infarction, but some waveforms that clinicians can easily diagnose existed that cannot be successfully detected by conventional programs. We let AI learn the detection of AF by convolutional neural network (CNN) using Lorenz plot of heart rate variability of 24-h ECG in subjects with sinus rhythm (SR) and AF whose diagnosis was confirmed as teacher data. Among 10000 datasets for SR and AF each, 80% of data was used as teacher data. With remaining 20% validation data, the CNN developed by AI detected AF with 100% sensitivity and 100% specificity.

Original languageEnglish
Title of host publication2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages177-180
Number of pages4
ISBN (Electronic)9781538663097
DOIs
Publication statusPublished - 2018 Dec 12
Externally publishedYes
Event7th IEEE Global Conference on Consumer Electronics, GCCE 2018 - Nara, Japan
Duration: 2018 Oct 92018 Oct 12

Publication series

Name2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018

Other

Other7th IEEE Global Conference on Consumer Electronics, GCCE 2018
Country/TerritoryJapan
CityNara
Period18/10/918/10/12

Keywords

  • Artificial intelligen (AI)
  • Atrial fibrillation (AF)
  • Convolutional neural network (CNN)
  • Holter electrocardiogram (ECG)
  • Lorenz plot

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Electrical and Electronic Engineering
  • Safety, Risk, Reliability and Quality
  • Instrumentation

Fingerprint

Dive into the research topics of 'Neural Network Detection of Atrial Fibrillation by Lorenz Plot Images of Interbeat Interval Variation'. Together they form a unique fingerprint.

Cite this