Scribd Logo
Upload

Welcome to Scribd!

  • Classification of Ecg Beats Using Two Dimensional Convolutional Neural Network To Detect Arrhythmia

    Uploaded by

    Nilesh Jadhav
    22 views18 pages
    This document summarizes a study that used a 2D convolutional neural network to classify 8 different types of ECG arrhythmia beats from the MIT-BIH Arrhythmia Database. The CNN model was trained on 128x128 pixel grayscale images of ECG signals and achieved an average accuracy of 99.05% and average sensitivities over 97.85% when classifying normal beats, premature ventricular contractions, right/left bundle branch blocks, and other arrhythmia types. The authors concluded the CNN approach was effective for automatic ECG arrhythmia detection and could potentially be applied in medical monitoring devices.

    Original Description:

    Conference paper published by me

    Original Title

    TPP-1

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    This document summarizes a study that used a 2D convolutional neural network to classify 8 different types of ECG arrhythmia beats from the MIT-BIH Arrhythmia Database. The CNN model was trained on 128x128 pixel grayscale images of ECG signals and achieved an average accuracy of 99.05% and average sensitivities over 97.85% when classifying normal beats, premature ventricular contractions, right/left bundle branch blocks, and other arrhythmia types. The authors concluded the CNN approach was effective for automatic ECG arrhythmia detection and could potentially be applied in medical monitoring devices.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    22 views18 pages

    Classification of Ecg Beats Using Two Dimensional Convolutional Neural Network To Detect Arrhythmia

    Uploaded by

    Nilesh Jadhav
    This document summarizes a study that used a 2D convolutional neural network to classify 8 different types of ECG arrhythmia beats from the MIT-BIH Arrhythmia Database. The CNN model was trained on 128x128 pixel grayscale images of ECG signals and achieved an average accuracy of 99.05% and average sensitivities over 97.85% when classifying normal beats, premature ventricular contractions, right/left bundle branch blocks, and other arrhythmia types. The authors concluded the CNN approach was effective for automatic ECG arrhythmia detection and could potentially be applied in medical monitoring devices.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    of 18

    Department of Electronics

    and Telecommunication

    CLASSIFICATION OF ECG
    BEATS USING TWO
    DIMENSIONAL
    CONVOLUTIONAL
    NEURAL NETWORK TO
    -GROUP 19:
    DETECT ARRHYTHMIA Ankita Patil (2015120034)
    Nilesh Jadhav (2016120076)
    Under the guidance of Prof.
    Anand Mane
    INTRODUCTION
    ● Cardiovascular Diseases cause 31% of deaths - World
    Health Organization (WHO)
    ● Arrhythmia refers to any irregular change from the
    normal heart rhythms
    ● Necessary to periodically monitor heart rhythms
    ● ECG displays rhythm and status of heart
    TYPES OF ARRHYTHMIA ECG BEATS TO CLASSIFY
    ● Normal beat (NOR)
    ● Paced beat (PAB)
    ● Ventricular escape beat (VEB)
    ● Premature ventricular contraction beat (PVC)
    ● Right bundle branch block beat (RBB)
    ● Atrial premature contraction beat (APC)
    ● Left bundle branch block beat (LBB)
    ● Ventricular flutter wave beat (VFW)
    PROCEDURE FOR CLASSIFICATION
    ECG ARRHYTHMIA CLASSIFIER
    EXPERIMENTAL INFORMATION
    ● MIT-BIH Arrhythmia Database
    ● ECG signals → 128 X 128 gray scale images
    ● Classification of eight types of Arrhythmia beats
    ● CNN model deployed in Python with TensorFlow
    ● Results displayed on Local Host
    DATA AUGMENTATION
    EVALUATION METRICS
    Accuracy(%) = TP+TN X 100 Sensitivity(%) = TP

    TP+TN+FP+FN TP+FN

    Specificity(%) = TN X 100 Positive Predictive Value(%)

    FP+TN = TP X100
    -True Positive (TP) : Correctly detected as arrhythmia
    TP+FP
    – True Negative (TN) : Correctly detected as normal

    – False Positive (FP) : Incorrectly detected as arrhythmia

    – False Negative (FN) : Incorrectly detected as normal


    LOCAL HOST WINDOW - CHOICE OF SAMPLE
    LOCAL HOST WINDOW - DISPLAYING NORMAL BEAT
    LOCAL HOST WINDOW -DISPLAYING PVC BEAT
    COMPARISON WITH PREVIOUS WORKS
    ANALYSIS AND CONCLUSION
    ● 99.05% Average Accuracy
    ● 99.57% Specificity
    ● 97.85% Average Sensitivity
    ● 98.55% Average Positive Predictive Value
    ● Effective approach
    ● Can be applied to the medical robot or the scanner that
    can monitors the ECG signals
    REFERENCES
    [1] World Health Organization (2017). Cardiovascular diseases (CVDs). [Online] Available: https://www.who.int/en/news-room/fact-sheets/
    detail/cardiovascular-diseases-(cvds). [Accessed: 18- Oct- 2018

    ] [2] Tae Joon Jun, Hoang Minh Nguyen, Daeyoun Kang, Dohyeun Kim, Daeyoung Kim, Young-Hak Kim, ’ECG arrhythmia classification using a 2-D
    convolutional neural network’

    [3] Linh TH, Osowski S, Stodolski M (2003).’On-line heart beat recognition using Her- mite polynomials and neuro-fuzzy network’ IEEE Transactions on
    Instrumentation and Measurement 52(4):1224-1231

    [4] Ceylan R, zbay Y ’Comparison of FCM, PCA and WT techniques for classification ECG arrhythmias using artificial neural network’ Expert Systems with
    Applications 33(2):286-295

    [5] Haseena HH, Mathew AT, Paul JK ’ Fuzzy clustered probabilistic and multi layered feed forward neural networks for electrocardiogram arrhythmia
    classification’ Journal of Medical Systems 35(2):179-188

    [6] Yu SN, Chen YH ’Electrocardiogram beat classification based on wavelet transformation and probabilistic neural network’ Pattern Recognition Letters
    28(10):1142-1150

    You might also like