RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication
Artículo científico (JournalArticle)
DYNAMO Framework: Advancing non-invasive, rapid calibration in cardiac digital twin technology
[EN] Introduction: Atrial arrhythmias management remains a significant challenge in cardiac electrophysiology. Cardiac digital twins (CDTs) offer a promising solution for individualized treatment. However, existing models are
limited by invasive input requirements or long computational times. This paper introduces DYNAMO, a
framework that leverages non-invasive electrocardiographic imaging (ECGI) to create CDTs.
Methods: DYNAMO utilizes a reaction-diffusion automata model, reducing computational demands while
capturing personalized cardiac dynamics using ECGI data. The calibration process maps ECGI data into CDTs by
extracting key parameters for personalization, specifically LATs and conduction velocities. The framework was
validated against detailed models and patient data in sinus rhythm (SR) and atrial flutter (AFl), comparing local
activation times (LATs) and body surface potential maps (BSPM) for models, and focusing on BSPM for patients.
Results: In-silico and clinical validation demonstrated the effectiveness of the framework. CDTs closely replicated
cardiac dynamics, with an average LAT difference of 3.8 ms for SR models and 8.03 ms for AFl. For patients,
CDTs achieved a high BSPM cross-correlation, with values averaging 0.89 for SR and 0.75 for AFl, and successfully replicated the ablation outcome in an AFl scenario. The framework also demonstrated superior
computational efficiency compared to traditional models for equivalent simulations.
Conclusions: DYNAMO represents a significant advancement in personalized cardiac health management. Using
non-invasive ECGI to create CDTs offers an efficient approach to cardiac modeling. Its capability to simulate
treatment strategies holds promise for improving clinical outcomes in atrial arrhythmia management. This study
lays the basis for the broader application of ECGI in personalized medicine., This work was supported by the Generalitat Valenciana Grants
(ACIF/2021/205, CIBEFP/2022/9, CIAPOS/2021/238, CIGE/2022/02,
CIAICO/2022/020) and by ESF Invest in your future. This work was
supported by MCIN/AEI/10.13039/501100011033 grants PID2020-
119364RB-I00, RYC2018-024346-I, CNS2022-135512, CPP2021-
008562, PTQ2023-013018, PTQ2023-012972 and by European Union
NextGenerationEU/PRTR and by FEDER Fondo Europeo de Desarrollo
Regional.
This research has received funding from the European Institute of
Innovation and Technology (EIT) under grant agreements No 220385
and 250027. This European body receives support from the Horizon
2020 research and innovation programme.
Project:
Agencia Estatal de Investigación / Plan Estatal de Investigación Científica y Técnica y de Innovación 2017-2020 PID2020-119364RB-I00
,
Agencia Estatal de Investigación / Plan Estatal de Investigación Científica y Técnica y de Innovación 2021-2023 CPP2021-008562
,
Generalitat Valenciana ACIF/2021/205
,
Generalitat Valenciana CIBEFP/2022/9
,
Generalitat Valenciana CIAICO/2022/020
,
Agencia Estatal de Investigación RYC2018-024346-I
,
Agencia Estatal de Investigación CNS2022-135512
,
EIT Health GA220385
,
European Commission 230129
,
Generalitat Valenciana PTQ2023-012972
,
Generalitat Valenciana CIAPOS/2021/238
,
Agencia Estatal de Investigación CIGE/2022/02
,
PTQ2023-013018