Researchers at IISER Pune have developed a novel way to analyze heart dynamics that could potentially be used in the diagnosis of abnormalities of the heart.
The analysis is based on the mathematical concept of fractals, which are repeating patterns that are similar at any scale. Fractals are commonly observed in nature, as in the branching patterns of trees or in the spiral patterns of sea shells. Researchers have since long used fractals as frameworks to define and describe complex systems.
The pattern of dynamics displayed by ECG (electrocardiograph) signals, which are a read-out for heartbeat, can be studied through fractal analysis owing to the non-uniform distribution of points in the underlying dynamics. Under a SERB project funded by DST, Prof. G. Ambika at IISER Pune along with her collaborator, Dr. K.P. Harikrishnan of Cochin College, led the research on the characterization of heart dynamics to differentiate between healthy and abnormal ECG signals.
By recreating the dynamics of the heart using derived measures from multifractal spectrum of ECG signals, the team has been able to successfully categorize healthy and unhealthy data sets. The study used a supervised machine learning approach to build a model to make these predictions with high accuracy.
Results from this work also indicate that in spite of high complexity, dynamics of healthy heart display less variability. The team believes that this method could lead to a quantitative way of analyzing ECG signals for diagnostics, therapy and continued monitoring of patients.
This study titled Detecting abnormality in heart dynamics from multifractal analysis of ECG signals is authored by Snehal M. Shekatkar, Yamini Kotriwar, K.P. Harikrishnan and G. Ambika and is published in Scientific Reports (7: 15127 (2017); DOI:10.1038/s41598-017-15498-z).
– Reported by Shanti Kalipatnapu; Graphic by Shubhankar Kulkarni