Detection of Myocardial Ischaemia based on Artificial Neural Networks and Skin Sympathetic Nerve Activity

EasyChair Preprint 2165

Abstract

In this study, we propose a new technique which detects the anomalies in skin sympathetic nerve activity (SKNA) recorded from the chest wall by using the state-of-the-art signal processing and machine learning methods for the robust detection of myocardial ischaemia (AMI). For this purpose, a preprocessing technique that obtains SKNA from the wideband recordings on STAFF III database, which are non-invasively recorded from the skin surface of the chest wall by using an equipment that has a wide frequency bandwidth and high sampling rate, is developed. By using the data that is obtained as a result of preprocessing, a novel feature extraction technique which obtains SKNA features that are critical for the reliable detection of AMI is developed. By using the critical SKNA features, a supervised learning technique based on artificial neural networks (ANN) which performs the robust detection of AMI is developed. The performance results of the proposed technique obtained from a considerable number of patients with coronary artery disease on STAFF III database indicate that the technique provides highly reliable detection of AMI.

Keyphrases: Artificial Neural Networks, Back propagation algorithm, Classification, ECG, Myocardial ischaemia, Sympathetic nerve activity, anomaly detection, coronary artery disease, feature extraction, skin sympathetic nerve activity

@booklet{EasyChair:2165,
  author    = {Merve Begüm Terzi and Makbule Kübra Korkmaz and Orhan Arıkan and Salih Topal and Adnan Abaci},
  title     = {Detection of Myocardial Ischaemia based on Artificial Neural Networks and Skin Sympathetic Nerve Activity},
  howpublished = {EasyChair Preprint 2165},
  year      = {EasyChair, 2019}}