ECG Image dataset for Myocardial Infarction Complication
Description
The ECG Image Dataset for Myocardial Infarction Complication is a curated collection of annotated electrocardiogram (ECG) images aimed at facilitating machine learning and deep learning research in the automated diagnosis of myocardial infarction (MI) and its related complications. This dataset primarily focuses on visual ECG data obtained from real-world clinical environments, including hospital ECG monitors and publicly available ECG databases. Each image in the dataset represents a single ECG recording, captured as a high-resolution image from clinical ECG equipment. The images have been preprocessed for clarity, standardized in resolution, and labeled by certified cardiologists or expert clinicians based on the morphological and waveform features indicative of various types of MI and cardiac complications. ECG images capture from ECG machine tuned on frequency 0.67- 25HZ and 10mm/mV lead voltage.
Files
Steps to reproduce
1. ECG Image Collection (ECG screen images collected from private clinics in Bhopal city region) 2. Preprocess images to crop, resize, and image noise reduction. 3. Medical experts suggestion to annotate ECGs into proper MI complication (medical expert / practitioners). 4. Organize images and labelling (Manual renaming).
Institutions
- VIT Bhopal University