Lead Concatenation in ECG Classification Using CWT: Required or Optional?

Dear Matlab Community,

I am currently working on a classification task with ECG recordings stored in a CSV file with dimensions of (5001, 12). The first row contains headers, and each column represents a lead of the ECG, totaling 12 leads. These recordings were made over a duration of 10 seconds at a sampling frequency of 500 Hz. Therefore, each lead comprises a sequence of 5000 values. The unit of measurement is 0.01 mV, adhering to the Philips standard recording system.

My specific question pertains to the methodology of feature extraction for classification purposes using the Continuous Wavelet Transform (CWT). Should I proceed by creating a scalogram for each lead independently and then concatenating them, or should I generate 12 scalograms from each ECG and consider them as belonging to the same class?

Put differently, is each lead's scalogram regarded as a class instance, or is it pivotal to concatenate the 12 leads to accurately represent an ECG? Your insights and guidance on this matter would be greatly appreciated.

Sincerely,

Answer 1

[William Rose]

@Sabrine,

I assume you have read this paper, published in 2021: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7831114/

The authors use only lead II for their classification algorithm. Lead II is typically used for a monitor display, when only a single lead is shown. I suggest you start with lead II only.

You ask "is each lead's scalogram regarded as a class instance, or is it pivotal to concatenate the 12 leads to accurately represent an ECG". I have not worked with classificaiton algorithms. Therefore please excuse me if I say things that are obvious to you, but not obvious to me. I may also say things that you already know. I am not trained in reading ECGs.

The different leads carry different information. Certain pathologies are more apparent in certain leads. Some pathologies display unique combinations of features in particular leads with a particular temporal relationship. For example, in torsades des pointes, there is a distinctive temporal progression of oscillations from one lead to another lead. Another example is ST segment elevation myocardial infarction (STEMI). The infarct may be classified according to which leads show the greatest ST segment elevation, and the diagnosis of STEMI is made with greater confidence when the electrically oppostite leads show reciprocal ST depression (see here). The feature extraction and learning algorithm should account for these sorts of phenomena.

The different leads may be thought of as projections of the instantaneous dipole moment of the heart through the body tissues onto the surface of the body. This is an overfsimplification, since the charge distribution in the heart at each instant is more complicated than a simple dipole. The electrical properties of the tissues, which affect how the dipole moment prohjects to the body surface, are complicated and change with time, due to breathing, pulse-related blood volume changes, muscle contraction, etc. I mention these facts just to make the point that the different leads of the ECG are distinct.

[Sabrine]

Thank you for your detailed insights and suggestions,@William Rose. I appreciate your thoughtful analysis of the paper and your explanations regarding the significance of different ECG leads.

[William Rose]

@Sabrine, you're welcome. Choose a representation that will allow the classifier to learn about patterns in which specific features occure uniquely on specific leads with particular temporal relationships.

If you are not already using them, check out the online databases:

MIMIC-IV-ECG database: >800,000 12-lead ECGs from >160,000 patients, 10 seconds each

PTB-XL: over twenty thoudsand 12-lead recordings, 10 seconds each

and other ECG databases at physionet.org.

[Sabrine]

Thank you @William Rose for the recommendations, Your guidance is greatly appreciated.

[William Rose]

You're welcome. Good luck.