A sensitive algorithm for automatic detection of space-time alternating signals in cardiac tissue

Area of Science:

• Cardiology
• Computational Biology
• Biophysics

Background:

• Cardiac alternans, beat-to-beat variations in cardiac signals, can precede life-threatening arrhythmias like ventricular tachycardia and fibrillation.
• Current methods for detecting alternans are often qualitative and ambiguous, especially in long-term, noisy datasets, hindering automated detection in large spatiotemporal data.

Purpose of the Study:

• To develop a quantitative definition and a novel algorithm for the automatic detection of cardiac alternans in the presence of noise.
• To enable early prediction and detection of arrhythmias by overcoming limitations in current alternans detection methods.

Main Methods:

• A combinatorics-derived quantitative definition of alternans was established.
• A new algorithm was developed utilizing temporal persistence (TP), representative phase (RP), and alternans ratio (AR) for automated detection.
• The algorithm was validated against theoretical probabilities and white noise test datasets.

Main Results:

• The novel algorithm successfully detected early, fine-scale calcium alternans in ultra-high resolution optical mapping data, even near the noise level.
• Detected alternans were linked to the subsequent formation of larger regions and the evolution of spatially discordant alternans (SDA).

Conclusions:

• The developed technique provides a robust method for quantifying alternans and understanding arrhythmia onset.
• This approach facilitates the analysis of space-time alternating signals and holds promise for improved arrhythmia prediction and detection.