Machine Learning Artificial Intelligence Early Detection Stroke Atrial Fibrillation

Project Information

MAESTRIA

Grant agreement ID: 965286

DOI

10.3030/965286

EC signature date 21 January 2021

Start date 1 March 2021

End date 31 August 2026

Funded under

• SOCIETAL CHALLENGES - Health, demographic change and well-being

Total cost

€ 13 915 311,25

EU contribution

€ 13 915 311,25

Coordinated by

SORBONNE UNIVERSITE
France

Periodic Reporting for period 2 - MAESTRIA (Machine Learning Artificial Intelligence Early Detection Stroke Atrial Fibrillation)

Reporting period: 2022-09-01 to 2024-02-29

Summary of the context and overall objectives of the project

Medical care of cardiac diseases needs to integrate more and more multiple parameters: functional and structural properties of cardiac tissues, genotype, metabolism or lifestyle. Multiple fields of expertise must be involved such as omics, biophysics of electrical signal, clinical imaging. Heterogeneous knowledge and data supporting complex and appropriate decisions to be taken by cardiologist clinicians for patient’s treatments; this new paradigm will revolutionise medical approaches and will durably impact the healthcare industry workflows starting from future emerging treatments to medical devices.

In Atrial Cardiomyopathy, characterised by a long period of clinically silent progression, and, most often, a dramatic acute clinical manifestation such as the new onset of symptomatic atrial fibrillation (AF), stroke or acute heart failure, the early detection of the disease and the recognition of precise underlying mechanisms are necessary for timely deployment of prevention strategies and personalised medical care.

The MAESTRIA project team focuses on developing novel approaches for timely detection of atrial myopathy to improve care management and identifying novel therapeutic targets for personalised medicine of AF and stroke. MAESTRIA addresses:
- Highly personalised diagnosis by combining research results on genomic, inflammation and metabolic disorders; all involved in the progression of atrial cardiomyopathy, incorporating AF and stroke in a novel, holistic and multidisciplinary patient health pathway. MAESTRIA has the potential of preventing excess deaths, stroke, and disability in 1.5-2% of the European population, and 12-15% of Europe’s octogenarians.
- Integration of data available from various sources by integrating mechanistic understanding of the cardiomyopathic substrate with cutting-edge imaging techniques, electrophysiological investigations and artificial intelligence approaches. Risk stratification in patients with AF will be refined, potential new therapeutic approaches to modify the natural history of AF and treatment rationalised.
- Development and deployment of a clinically applicable digital diagnosis platform by contributing to the development of a precision medicine tool set for the contemporary management of AF and stroke across Europe and beyond. MAESTRIA will build an open-access digital platform for clinicians and patients in Europe to inform them about risk based on the outcome of our investigations.

The MAESTRIA project team aims to develop and validate the first integrative diagnostic digital platform for atrial cardiomyopathy diagnosis. This platform will be designed to provide support for improved diagnostic accuracy that increases effectiveness and efficiency of treatments, as well as prevention of the complications of atrial cardiomyopathy, such as atrial fibrillation and stroke.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The goal of the MAESTRIA is to develop and validate the first integrative digital platform for the diagnostic of cardiomyopathy and to improve diagnostic accuracy, treatment effectiveness as well as prevention of atrial fibrillation and stroke, the two complications of the atrial cardiomyopathy.
The project is organized in eight WPs, from upstream research to deployment of applicable digital diagnostic platform and including ethics, valorisation, and network management.
• WP1, validation of multimodality imaging biomarkers. Progress has been achieved towards the sharing of clinical and scan data between partners to join the ORFAN Study to validate the atrial score. Implementation, training, and validation of a deep neural network for AI-based dynamic multimodality segmentation and atrial strain has been realized as well as the collection, and expert annotations of 1400 cine MRI loops of the heart and 2100 cine loops. A robust pseudonymisation tool is now available that will be used for the prospective study (WP4).
• WP2, electrophysiological parameters as biomarkers of the atrial cardiomyopathy. Progress have been made in linking molecular mechanisms of atrial cardiomyopathy to electrophysiological abnormalities by detailed analyzing of biological data of the Catch-ME cohort. The prospective clinical study, CATS AF designed to confront intra cardiac electrical mapping with MRI imaging has started with the recruitment of the first patients with de novo AF.
• WP3, the discovery of new biomarker of the atrial cardiomyopathy. The study of immune cell subpopulations involved in atrial cardiomyopathy is progressing well using single cell transcriptomic approach and experimental models of atrial cardiomyopathy (collaboration Boston, Paris). It is possible to analyze using PET imaging, DFA storage specifically in the atria opening now perspective to study the impact of metabolism on atrial functional properties. Epicardial adipose tissue is a source of inflammasome that could be a potential biomarker of the atrial cardiomyopathy.
• WP4, prospective patient cohort to validate multimodality biomarkers. The consortium agreement has been signed by the 18 partners. Processes for patient recruitment as well as the various core labs to centralize clinical data are now setup. Patient recruitment has started with a rate of inclusion of patients in line with the project mile stones.
• WP5, implementation of the MAESTRIA demonstrator. All legal, organizational, and operational aspects of the MAESTRIA demonstrator are setup. Fist multimodality analysis have been conducted using MRI and echocardiography imaging.
• WP 6, dissemination. The consortium has been well visible through its website and thanks to the number of articles published by MAESTRIA partner as well as their active participations to congresses, conferences and workshops.
• WP7, project management. Maestria is managed by a core management team and weekly meetings are held for this purpose. In addition, coordination between the work packages is ensured through monthly meetings to monitor the project's progress. Maestria has also so far organised two annual meetings with a large number of participants and with good results in terms of consortium building and organisation of work.
• WP8, Ethics requirement. A delay should be noted in submitting the corresponding deliverables. However, the ethical board is now fully operational and has reorganized the transfer of information from the partners of the consortium and from the WP leader on ethics items, which will facilitate the finalisation of the remaining deliverables

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

N/A

LOGO MAESTRIA