Periodic Reporting for period 2 - MeDiTATe (The Medical Digital Twin for Aneurysm Prevention and Treatment)
Reporting period: 2022-01-01 to 2024-06-30
Cardiovascular diseases (CVDs) are the main cause of illness and premature death in the EU. Personalized medicine and the use of simulations are the most promising areas of research with high potential benefits for patients, citizens and economy. The use of scientific computing tools in medical practice is, nevertheless, still in its infancy. The main causes are related to the requirement of elevated high-tech engineering know-how, the cost of computations and the poor numerical validation of tools against clinical studies. In this scenario, the MeDiTATe project aimed to deliver a comprehensive framework of simulation and imaging technologies, targeted at industrial and clinical-translation to accelerate the process of personalised cardiovascular medical procedures to ultimately improve patient care. The strength of the concept lies within the multi-disciplinary aspect given by the intense collaboration among different technical experts in the field of engineering, clinics, academics and industry. The core idea is to develop image based Medical Digital Twins of cardiovascular districts for a patient specific prevention and treatment of aneurysms and to make it available as "a service" for all in academia, hospital and industry.
The project involved 14 Individual Research Projects (IRP) defined across five research tracks:
1. High fidelity CAE multi-physics simulation with RBF mesh morphing (FEM, CFD, FSI, inverse FEM);
2. Real time interaction with the digital twin by Augmented Reality, Haptic Devices and Reduced Order Models;
3. HPC tools, including GPUs, and cloud-based paradigms for fast and automated CAE processing of clinical database;
4. Big Data management for population of patients imaging data and high-fidelity CAE twins;
5. Additive Manufacturing of physical mock-up for surgical planning and training to gain a comprehensive Industry 4.0 approach in a clinical scenario (Medicine 4.0).
The project has been successfully concluded, with 14 researchers completing their research paths and 10 obtaining PhD titles within 2024. Key outcomes of the projects include novel workflows for real-time hemodynamic assessment and aneurysm growth predictions, GPU-accelerated simulation tools, and uncertainty quantification models to enhance computational efficiency. The project also introduced advanced imaging techniques and auxetic stent graft designs for improved endovascular repair, mock loops for validating endovascular procedures, and non-invasive ultrasound-based diagnostics for assessing aneurysm mechanical properties.
The project involved 14 Individual Research Projects (IRP) defined across five research tracks:
1. High fidelity CAE multi-physics simulation with RBF mesh morphing (FEM, CFD, FSI, inverse FEM);
2. Real time interaction with the digital twin by Augmented Reality, Haptic Devices and Reduced Order Models;
3. HPC tools, including GPUs, and cloud-based paradigms for fast and automated CAE processing of clinical database;
4. Big Data management for population of patients imaging data and high-fidelity CAE twins;
5. Additive Manufacturing of physical mock-up for surgical planning and training to gain a comprehensive Industry 4.0 approach in a clinical scenario (Medicine 4.0).
The project has been successfully concluded, with 14 researchers completing their research paths and 10 obtaining PhD titles within 2024. Key outcomes of the projects include novel workflows for real-time hemodynamic assessment and aneurysm growth predictions, GPU-accelerated simulation tools, and uncertainty quantification models to enhance computational efficiency. The project also introduced advanced imaging techniques and auxetic stent graft designs for improved endovascular repair, mock loops for validating endovascular procedures, and non-invasive ultrasound-based diagnostics for assessing aneurysm mechanical properties.
The complex physics involved in the aneurysm solicitations require computationally expensive high fidelity (HiFi) analyses. Procedures aimed to investigate the Ascending Aorta Aneurism evolution and rupture are developing moving from the images of the vessel directly to the analysis of the simulation results. A prototype, which the clinicians will be able to use, has been developed. It allows to quickly obtain geometric measurements on the aortic model to monitor the progression of the pathology and exploit the full power of ROM to obtain real-time results, providing crucial information on the haemodynamic flow. With such tools the risk of rupture can be estimated by combining metrics on the geometry and fluid-dynamic biomarkers, ensuring a more patient-specific understanding of the problem and a strong incentive for the revolution of the clinical practice.
Clinical data from CT-scans and 2D/3D flows have been used to validate the simulations and create patient-specific aorta geometries. An idealized AAA was used as a benchmark for exploring the fitting and positioning of a stent graft device within the aortic wall. Computational simulations were focused on the exploration of innovative stent designs able to improve key performance parameters of existing commercial products. An innovative manufacturing workflow to produce a high-fidelity tissue mimicking transparent phantom of AAA was investigated.
Key achievements in the last period include the refinement of high-fidelity patient-specific cardiovascular simulations. Advanced Reduced Order Models (ROM) and Uncertainty quantification (UQ) methods were implemented. GPU-accelerated workflows enabled real-time predictions of aneurysm behaviour. New machine learning techniques were integrated into the modelling framework to enhance the accuracy of hemodynamic predictions.
Innovative 3D-printed models and experimental validation techniques were developed, supporting the translation of digital twin concepts into clinical practice. Studies on novel stent graft designs and auxetic structures provided insights into optimizing endovascular treatments. The combination of clinical imaging data with computational simulations led to the creation of automated workflows for assessing aneurysm progression and rupture risks.
The project results were shared through scientific publications and collaborative initiatives with clinical and industrial partners. The project was showcased at large number of conferences, significantly increasing its visibility. Beyond academic dissemination, the project also engaged in a variety of outreach activities designed to communicate the project’s aims and results to a broader audience. These included visits to schools, participation in Researchers' Night events, and other community engagement activities. Three patents were granted by different ESRs and two Med-Tech Start-ups involved: PrediSurge (www.predisurge.com) and LivGemini (www.livgemini.com). The results have been shared on the Horizon Results Platform.
Clinical data from CT-scans and 2D/3D flows have been used to validate the simulations and create patient-specific aorta geometries. An idealized AAA was used as a benchmark for exploring the fitting and positioning of a stent graft device within the aortic wall. Computational simulations were focused on the exploration of innovative stent designs able to improve key performance parameters of existing commercial products. An innovative manufacturing workflow to produce a high-fidelity tissue mimicking transparent phantom of AAA was investigated.
Key achievements in the last period include the refinement of high-fidelity patient-specific cardiovascular simulations. Advanced Reduced Order Models (ROM) and Uncertainty quantification (UQ) methods were implemented. GPU-accelerated workflows enabled real-time predictions of aneurysm behaviour. New machine learning techniques were integrated into the modelling framework to enhance the accuracy of hemodynamic predictions.
Innovative 3D-printed models and experimental validation techniques were developed, supporting the translation of digital twin concepts into clinical practice. Studies on novel stent graft designs and auxetic structures provided insights into optimizing endovascular treatments. The combination of clinical imaging data with computational simulations led to the creation of automated workflows for assessing aneurysm progression and rupture risks.
The project results were shared through scientific publications and collaborative initiatives with clinical and industrial partners. The project was showcased at large number of conferences, significantly increasing its visibility. Beyond academic dissemination, the project also engaged in a variety of outreach activities designed to communicate the project’s aims and results to a broader audience. These included visits to schools, participation in Researchers' Night events, and other community engagement activities. Three patents were granted by different ESRs and two Med-Tech Start-ups involved: PrediSurge (www.predisurge.com) and LivGemini (www.livgemini.com). The results have been shared on the Horizon Results Platform.
The innovative nature of the projects is expected to have a broader impact on the research environment of the beneficiaries, promoting best practice in research within institutions and enhancing engagement for industries and hospital research centres. The results of this research can be proposed as a guide for all companies wanting to implement the Medical Digital Twin (MDT) in cardiovascular application. The tools developed to predict the long-term evolution of a pathology and to assess the risk factors will allow translating this research into the clinical field, providing insights that can help clinicians in the evaluation and treatment planning, as part of medical digital twin and personalized medicine approaches.
With such results, MeDiTATe contributes to the development of high quality care solutions for individual patients and sensible management for cardiovascular chronic diseases. The research work carried out can be a real guide especially for all those who want to develop market approved software from this example carried out exclusively for research purposes. It can open new markets in neurovascular, reduce costs associated with heavy numerical computations, and open new business model thanks to results being obtained in real-time. The MeDiTATe research responds to a need from industrial partners, who depend on the guidance of clinical counterparts to direct R&D over the mid and long-term. Our clinical counterparts are at the forefront of translational research, thriving in innovation, being centres of excellence not only in their own countries but internationally, supportive of the use of novel technologies in healthcare.
From a societal point of view, the MeDiTATe research contributes to progress towards a reduction of overall mortality and surgery-related morbidity.
With such results, MeDiTATe contributes to the development of high quality care solutions for individual patients and sensible management for cardiovascular chronic diseases. The research work carried out can be a real guide especially for all those who want to develop market approved software from this example carried out exclusively for research purposes. It can open new markets in neurovascular, reduce costs associated with heavy numerical computations, and open new business model thanks to results being obtained in real-time. The MeDiTATe research responds to a need from industrial partners, who depend on the guidance of clinical counterparts to direct R&D over the mid and long-term. Our clinical counterparts are at the forefront of translational research, thriving in innovation, being centres of excellence not only in their own countries but internationally, supportive of the use of novel technologies in healthcare.
From a societal point of view, the MeDiTATe research contributes to progress towards a reduction of overall mortality and surgery-related morbidity.