Skip to Main content Skip to Navigation

Novel pre-interventional atrial flutter localization tool for the improvement of radiofrequency ablation efficacy

Abstract : Prevalence of atrial flutter (AFL) is predicted to increase in the coming years. AFL involves a rapid and regular activation of the atrium due to defects in the activation propagation, and predisposes to life-threatening conditions. Radiofrequency catheter ablation is a well-known AFL treatment option for its efficiency. Yet procedural efficacy is rather poor due to poor amounts and quality of information on right or left AFL localization available before invasive procedure. This memoir shows how to exploit the variability of AFL on the electrocardiogram (ECG) to localize AFL circuits non-invasively. Two original complimentary methodologies were developed based on (1) beat-to-beat serial vectorcardiographic (VCG) loop analysis, and (2) recurrence quantification analysis. For this to work, novel processing methods were developed for (a) flutter wave detection using likelihood-ratio tests, (b) correction of T wave overlaps using polynomial splines and respiratory motion using improved classical estimators, and (c) transformation into VCG using an optimized transform. Machine learning techniques such as feature selection and an original cross-validation approach were employed to induce practical linear binary classifiers with good localization performance without classifier overfitting. Furthermore, it was shown that variability of flutter waves are likely related to the circuit and not to sources of distortion such as respiratory motion or T wave. Relevant features allowed insight into the pathology, and showed that right AFL was much slower and more variable than left AFL.
Complete list of metadata

Cited literature [108 references]  Display  Hide  Download
Contributor : Abes Star :  Contact Connect in order to contact the contributor
Submitted on : Tuesday, June 9, 2020 - 3:08:09 PM
Last modification on : Wednesday, October 14, 2020 - 4:22:51 AM


Version validated by the jury (STAR)


  • HAL Id : tel-02862431, version 1



Muhammad Haziq Bin Kamarul Azman. Novel pre-interventional atrial flutter localization tool for the improvement of radiofrequency ablation efficacy. Signal and Image processing. COMUE Université Côte d'Azur (2015 - 2019); Universiti Kuala Lumpur (Malaisie), 2019. English. ⟨NNT : 2019AZUR4079⟩. ⟨tel-02862431⟩



Record views


Files downloads