Digitization of Paper Electrocardiogram: A Review

Digitization of Paper Electrocardiogram: A Review

Priyanka Gautam (Dr. B. R. Ambedkar National Institute of Technology Jalandhar, India), Ramesh Kumar Sunkaria (Dr. B. R. Ambedkar National Institute of Technology Jalandhar, India) and Lakhan Dev Sharma (MLV Textile and Engineering College, India)
Copyright: © 2019 |Pages: 17
DOI: 10.4018/978-1-5225-7525-2.ch009
OnDemand PDF Download:
No Current Special Offers


In order to diagnose a possible cardiac disorder, ECG (electrocardiogram) signals are usually recorded on standard grid papers in hospitals. Many efforts have been made to advance the technology in order to improve the diagnosis and management of cardiovascular disease. There is a need to convert the existing ECG records into digital forms, as it is the most efficient method to store and analyze ECG attributes for clinical uses. The main purpose of this chapter is to review the existing algorithms for digital conversion of paper ECG. It discusses the various challenges and a systematic study on different methods that have been used so far to convert paper ECG records into digitized form so that they can be retrieved efficiently. Initial challenge involved in the digitization process is gridline removal. In this process, information of ECG signal is also removed. None of the existing methods provide flawless gridline removal. The paper ECG used in hospitals differs in shape, size, formats, so the main challenge in digitization process is to achieve a worldwide ECG format.
Chapter Preview


Technologies that are used to compute human behavior or physical features are known as Biometrics such as iris, face, fingerprints, retina, hand geometry, voice or signatures and using such measures to detect and recognize individuals. Biometric act as an acrimonious alternative for identity verification such as passports, driving licenses, ID cards or PIN numbers in authentication. No need to bring any type of additional ID documents by the user because biometrics uses unique physical traits for authentication purpose. Biometric credentials cannot be forgotten, lost, easily cloned or guessed like other traditional authorization systems such as passwords, ID cards or personal identification numbers (PINs). Usual biometric systems involve a nomination and verification/identification phase. In enrollment process “live samples” of a person's biometric is acquired for identification purpose, followed by processing process and after that stored as templates. In verification process captured biometric samples are matched to the enrolled template, which is stored so as to verify/identify user identity (Lumini et al., 2017, Merone et al., 2017, Hejazi et al., 2016).

First fundamental fingerprint recognition system was presented in early 20 century. In search of new biometric modalities, research community spent too much energy to get any behavioral or physical trait which assure the conditions of universality, varying with time, discriminate between the population, readily collectable and difficult to cheat/reproduce. On the basis of above criteria several distinctive symptoms are identified, such as physiological (e.g. fingerprint, face, and iris), behavioral (e.g. signature, voice, gait), soft (e.g. gender, height and ethnicity) and medico-chemical (e.g. ECG, DNA) (Lumini et al., 2017, Sidek et al., 2014).

Unlike traditional biometric modalities, for authentication purpose the heart is likely a most safe for biometric modality since it is confined in the structure of the body, making it difficult for the opportunists to forge, change, replicate, mimic etc. The electric signal produced by the heart can be captured in a non-invasive way from the body's surface using ECG sensors. Most of the available devices for ECG acquisition in developing and underdeveloped countries use Paper-ECG. To create a country wide database for citizen identification/ verification, it is required to develop techniques for digitization of these ECG's available in paper form (i.e. Paper-ECG) (Islam et al., 2017, Gutta et al., 2016).

Key Terms in this Chapter

Systole: Contracting or pumping phase of a cardiac chamber.

Electrocardiogram: Electrical activity of the heart is recorded in the form of a waveform.

Diastole: Relaxing or filling phase of a cardiac chamber.

Histogram: Histogram of an image plots the number of pixels for each intensity level.

Depolarization: An excited cell which shows action potential is known as depolarized and the process is called depolarization.

Repolarization: After getting into the depolarized state for a definite interval of time, the cells are again polarized and go back into resting potential state through repolarization process.

Vectorcardiography: It is the technique to examine the electrical activity of heart by acquiring ECG along three axes.

Complete Chapter List

Search this Book: