The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Récemment, le système de diagnostic basé sur l'ECG et les appareils portables a attiré de plus en plus l'attention des chercheurs. Les études existantes ont atteint une précision de classification élevée en utilisant des réseaux neuronaux profonds (DNN), mais certains problèmes subsistent, tels que : une segmentation imprécise des battements cardiaques, une utilisation inadéquate des connaissances médicales, le même traitement de caractéristiques d'importance différente. Pour résoudre ces problèmes, cet article : 1) propose une méthode adaptative de segmentation-remodelage pour acquérir de nombreux échantillons utiles ; 2) construit un ensemble de fonctionnalités artisanales et de fonctionnalités approfondies sur l'échelle des battements internes, des battements et des inter-battements en intégrant suffisamment de connaissances médicales. 3) a introduit un module d'attention de canal (CAM) modifié pour augmenter les canaux importants dans les fonctionnalités profondes. Suite à la recommandation de l'Association for Advancement of Medical Instrumentation (AAMI), nous avons classé l'ensemble de données en quatre classes et validé notre algorithme sur la base de données MIT-BIH. Les expériences montrent que la précision de notre modèle atteint 96.94 %, soit une augmentation de 3.71 % par rapport à celle d'une alternative de pointe.
Huan SUN
Beijing Jiaotong University
Yuchun GUO
Beijing Jiaotong University
Yishuai CHEN
Beijing Jiaotong University
Bin CHEN
Beijing Jiaotong University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Huan SUN, Yuchun GUO, Yishuai CHEN, Bin CHEN, "ECG Classification with Multi-Scale Deep Features Based on Adaptive Beat-Segmentation" in IEICE TRANSACTIONS on Communications,
vol. E103-B, no. 12, pp. 1403-1410, December 2020, doi: 10.1587/transcom.2020SEP0002.
Abstract: Recently, the ECG-based diagnosis system based on wearable devices has attracted more and more attention of researchers. Existing studies have achieved high classification accuracy by using deep neural networks (DNNs), but there are still some problems, such as: imprecise heart beat segmentation, inadequate use of medical knowledge, the same treatment of features with different importance. To address these problems, this paper: 1) proposes an adaptive segmenting-reshaping method to acquire abundant useful samples; 2) builds a set of hand-crafted features and deep features on the inner-beat, beat and inter-beat scale by integrating enough medical knowledge. 3) introduced a modified channel attention module (CAM) to augment the significant channels in deep features. Following the Association for Advancement of Medical Instrumentation (AAMI) recommendation, we classified the dataset into four classes and validated our algorithm on the MIT-BIH database. Experiments show that the accuracy of our model reaches 96.94%, a 3.71% increase over that of a state-of-the-art alternative.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2020SEP0002/_p
Copier
@ARTICLE{e103-b_12_1403,
author={Huan SUN, Yuchun GUO, Yishuai CHEN, Bin CHEN, },
journal={IEICE TRANSACTIONS on Communications},
title={ECG Classification with Multi-Scale Deep Features Based on Adaptive Beat-Segmentation},
year={2020},
volume={E103-B},
number={12},
pages={1403-1410},
abstract={Recently, the ECG-based diagnosis system based on wearable devices has attracted more and more attention of researchers. Existing studies have achieved high classification accuracy by using deep neural networks (DNNs), but there are still some problems, such as: imprecise heart beat segmentation, inadequate use of medical knowledge, the same treatment of features with different importance. To address these problems, this paper: 1) proposes an adaptive segmenting-reshaping method to acquire abundant useful samples; 2) builds a set of hand-crafted features and deep features on the inner-beat, beat and inter-beat scale by integrating enough medical knowledge. 3) introduced a modified channel attention module (CAM) to augment the significant channels in deep features. Following the Association for Advancement of Medical Instrumentation (AAMI) recommendation, we classified the dataset into four classes and validated our algorithm on the MIT-BIH database. Experiments show that the accuracy of our model reaches 96.94%, a 3.71% increase over that of a state-of-the-art alternative.},
keywords={},
doi={10.1587/transcom.2020SEP0002},
ISSN={1745-1345},
month={December},}
Copier
TY - JOUR
TI - ECG Classification with Multi-Scale Deep Features Based on Adaptive Beat-Segmentation
T2 - IEICE TRANSACTIONS on Communications
SP - 1403
EP - 1410
AU - Huan SUN
AU - Yuchun GUO
AU - Yishuai CHEN
AU - Bin CHEN
PY - 2020
DO - 10.1587/transcom.2020SEP0002
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E103-B
IS - 12
JA - IEICE TRANSACTIONS on Communications
Y1 - December 2020
AB - Recently, the ECG-based diagnosis system based on wearable devices has attracted more and more attention of researchers. Existing studies have achieved high classification accuracy by using deep neural networks (DNNs), but there are still some problems, such as: imprecise heart beat segmentation, inadequate use of medical knowledge, the same treatment of features with different importance. To address these problems, this paper: 1) proposes an adaptive segmenting-reshaping method to acquire abundant useful samples; 2) builds a set of hand-crafted features and deep features on the inner-beat, beat and inter-beat scale by integrating enough medical knowledge. 3) introduced a modified channel attention module (CAM) to augment the significant channels in deep features. Following the Association for Advancement of Medical Instrumentation (AAMI) recommendation, we classified the dataset into four classes and validated our algorithm on the MIT-BIH database. Experiments show that the accuracy of our model reaches 96.94%, a 3.71% increase over that of a state-of-the-art alternative.
ER -