ABSTRACT

Atrial tachycardia (AT) and fibrillation (AF) are common cardiac arrhythmias that affect >2% of the global population [1]. They are characterised by rapid electrical activations in the atria, which are commonly due to the underlying pathological changes of cardiac tissue – fibrosis [2]. AT and AF can be detected in a patient from unusual morphologies of patient body-surface electrocardiogram (ECG). However, patients often develop fibrotic cardiomyopathy before the arrhythmias, making their detection from ECG very challenging. Early detection of subtle ECG changes due to the underlying cardiomyopathies can enable efficient prevention of the arrhythmias. This project aims to develop an efficient AI-based tool to identify arrhythmic P-waves from ECG and to predict the underlying arrhythmia types (AT/AF) and cardiomyopathy. This overarching aim will be achieved through the following objectives: 1) simulate normal and AT/AF electrical activations in 3D upper body and atrial models with fibrosis – patient’s digital twins [3-8]; 2) calculate the respective 12-lead ECG signals in the digital twins; 3) apply the ECG data to train AI to predict the type of arrhythmia and underlying cardiomyopathy; 4) test the AI on two unseen sets of ECG recordings, one from the digital twins and one from patients. Thus, the project will create a novel AI tool that can predict the arrhythmias and the underlying fibrotic cardiomyopathy, even during the normal rhythm. The early detection and prevention of AT and AF can greatly ease the burden on healthcare systems.

REFERENCES

[1] Faust O, Ciaccio EJ, Acharya UR. (2020). Review of Atrial Fibrillation Detection Methods as a Service. Int J Environ Res Public Health. 17(9): 3093.

[2] Xintarakou A, Tzeis S, Psarras S, et al. (2020) Atrial fibrosis as a dominant factor for the development of atrial fibrillation: facts and gaps. EP Europace 22(3), 342–351.

[3] Colman MA, Aslanidi OV, Stott J, et al. (2011). Correlation between P-wave morphology and origin of atrial focal tachycardia. IEEE Trans Biomed Eng. 58 (10), 2952-2255.

[4] Aslanidi OV, Colman MA, Stott J, et al. (2011). 3D virtual human atria: A computational platform for studying clinical atrial fibrillation. Prog Biophys Mol Biol 107 (1), 156-168.

[5] Muffoletto M, Qureshi A, …, Aslanidi O. (2021). Toward patient-specific prediction of ablation strategies for atrial fibrillation using deep learning. Front Physiol. 12, 674106.

[6] Ogbomo-Harmitt S, Muffoletto M, , …, Aslanidi O (2023) Exploring interpretability in deep learning prediction of successful ablation therapy for atrial fibrillation. Front. Physiol. 14:1054401. 

[7] Corral-Acero J, Margara F, … Lamata P. (2020) The 'Digital Twin' to enable the vision of precision cardiology. Eur Heart J. 41(48): 4556-4564. 

[8] Lamata P, Casero R, Caparella V, et al. (2014) Images as drivers of progress in cardiac computational modelling. Prog Biophys Mol Biol. 115(2-3): 198–212.