ECG Paper Record Digitization and Diagnosis Using Deep Learning

Abstract

Purpose Electrocardiogram (ECG) is one of the most essential tools for detecting heart problems. Till today most of the ECG records are available in paper form. It can be challenging and time-consuming to manually assess the ECG paper records. Hence, automated diagnosis and analysis are possible if we digitize such paper ECG records. Methods The proposed work aims to convert ECG paper records into a 1-D signal and generate an accurate diagnosis of heart-related problems using deep learning. Camera-captured ECG images or scanned ECG paper records are used for the proposed work. Effective pre-processing techniques are used for the removal of shadow from the images. A deep learning model is used to get a threshold value that separates ECG signal from its background and after applying various image processing techniques threshold ECG image gets converted into digital ECG. These digitized 1-D ECG signals are then passed to another deep learning model for the automated diagnosis of heart diseases into different classes such as ST-segment elevation myocardial infarction (STEMI), Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), and T-wave abnormality. Results The accuracy of deep learning-based binarization is 97%. Further deep learning-based diagnosis approach of such digitized paper ECG records was having an accuracy of 94.4%. Conclusions The digitized ECG signals can be useful to various research organizations because the trends in heart problems can be determined and diagnosed from preserved paper ECG records. This approach can be easily implemented in areas where such expertise is not available. Supplementary Information The online version contains supplementary material available at 10.1007/s40846-021-00632-0.

The History and Challenges of SCP-ECG: The Standard Communication Protocol for Computer-Assisted Electrocardiography

“…Paradoxically, while we are already in the digital era, in several use cases, these "ECG images" need to be processed shortly after being printed, to recover the "voltage-versustime" nature of the signals [104][105][106] (this was the subject of P.R. 's PhD thesis in the late 1960s!…”

“…Some authors claim that by using this "reverse-printing" process they can achieve (only) up to 97% digitization accuracy [105], a figure that does not guarantee that the reconstructed signal will allow detecting small Q waves as defined in [13]. Other AI researchers further argue that the original ECG signals are not really needed as AI-based ECG interpretation may be performed by directly processing the "ECG images", just as any medical image record [103,106,107]. However, these new approaches have not yet been assessed against well-documented databases such as the CSE Diagnostic database.…”

Hybrid Detection Model for Meticulous Presaging of Heart Disease using Deep Learning: HDMPHD

“…For ECG achieved 94 % accuracy. Han C et.al [6], this study presents an AI-based model in which demonstrates the multiple leads based on acute myocardial infarction with asynchronous ECG lead sets the developing algorithm on smart watches. This algorithm for smart watches easily detects cardiac disorders.…”

“…For a few instants, another peak is labeled with U. The upper peak is labeled with U and the heart chambers are denoted by P wave showing atria depolarization, the lower heart chamber is denoted by T wave and the QRS complex represents depolarization in ventricles [6] as shown in Figure 1.…”

Development and Validation of an Algorithm for the Digitization of ECG Paper Images

“…Furthermore, saving the patient’s history guarantees the ease of knowing his clinical evolution. Finally, knowing the entire time series of each signal, it is possible to implement some algorithms for the automatic detection of pathologies [ 1 , 2 , 3 ].…”