Parkinson’s disease affects more than one million people in the EU and this number is expected to double by 2030, primarily due to an aging population.
To mark World Parkinson’s Day, HaDEA interviewed Prof. Leontios Hadjileontiadis, coordinator of AI-PROGNOSIS, a Horizon Europe research and innovation project aiming to advance Parkinson’s disease diagnosis and care through novel predictive models combined with digital biomarkers from everyday devices, such as smartphones and smartwatches.
Prof. Hadjileontiadis, tell us more about AI-PROGNOSIS.
AI-PROGNOSIS is focused on improving Parkinson’s disease diagnosis and care through predictive models driven by artificial intelligence (AI) and digital biomarkers from everyday devices. The project aims to enhance early detection, predict disease progression and optimise treatment responses, thereby personalising patient care. By leveraging data from smartphones and smartwatches, AI-PROGNOSIS offers valuable insights into individual risk and treatment efficacy, ultimately improving the quality of life for those with Parkinson’s disease.
What can you tell us about your project’s use of AI? Have you encountered any challenges in integrating your solutions in broader healthcare systems?
The lack of interoperability with legacy electronic health record systems has made integrating AI tools into existing clinical workflows difficult. Many healthcare institutions still use outdated systems not designed to support advanced AI technologies. Building trust and acceptance among healthcare professionals and patients has also been a challenge. There is often scepticism about the accuracy and reliability of AI models and concerns about the potential for AI to replace human roles in healthcare. Overcoming these concerns requires continuous education and demonstration of the AI tools’ benefits and reliability.
Navigating the complex regulatory landscape for AI in healthcare has added to the challenges. Ensuring that AI tools meet all legal and ethical standards is essential for their adoption and use. Additionally, accessing existing datasets has been difficult due to data ownership and sharing restrictions, which limit the amount of data available for training AI models. Recruiting patients for studies and trials has also been challenging, as it requires significant time and resources to ensure a diverse and representative sample.
These challenges underscore the importance of a collaborative and adaptive approach in developing and implementing AI solutions in healthcare, ensuring they are both effective and widely accepted.
Could you elaborate on this collaborative and adaptive approach?
AI-PROGNOSIS has adopted a comprehensive and inclusive approach to identify the needs of key stakeholders, including patients, healthcare professionals, and researchers. The project emphasises continuous engagement and collaboration with these groups to ensure the tools developed are user-friendly and meet their needs.
This includes:
Multidisciplinary workshops: AI-PROGNOSIS organises workshops bringing together experts from various fields to discuss and refine project goals and methodologies;
Patient involvement: Patients are actively involved in the design and testing phases, providing valuable feedback on usability and functionality;
Input from health professionals: Regular consultations with doctors and therapists help tailor the AI tools to clinical workflows and practical needs;
Input from the external advisory board: Expert guidance on the ethical implementation, strategic integration, industry perspectives and impactful application of AI-PROGNOSIS output in Parkinson’s disease research and care.
This collaborative and iterative approach ensures that AI-PROGNOSIS remains aligned with the real-world needs of its stakeholders, enhancing its impact on Parkinson’s diagnosis and care.
Having consulted with numerous stakeholders, can you give us an example of how this feedback has been used?
For example, in one of the focus groups that we ran, healthcare professionals shared how challenging it was to be informed about their patients’ changing symptoms across the course of the illness. This insight helped us develop the mAI-Insights application, which allows healthcare professionals to receive frequent updates and alerts about their patients’ symptoms.*
With the project running until 2027, how important is the EU’s financial support throughout the project life cycle?
The support of EU funding is crucial for our project. It provides financial resources for extensive research, developing advanced AI models, and integrating digital biomarkers from everyday devices. EU funding under the Horizon Europe programme also facilitates collaboration among multidisciplinary European teams, ensuring that the project benefits from diverse expertise and perspectives. Additionally, this support helps navigate regulatory challenges and promotes the adoption of innovative solutions in healthcare systems. Without EU funding, achieving the project’s ambitious goals and significantly impacting Parkinson’s diagnosis and care would be much more challenging.
*The project also features two other applications: mAI-Health for persons with suspected Parkinson’s to track their personalised risk and mAI-CARE for persons with diganosed Parkinson’s to track symptoms, disease progression and treatment efficacy.
Background
Horizon Europe is the research and innovation programme of the EU for the period 2021-2027. The aims of Cluster 1 ‘Health’ include improving and protecting the health and well-being of citizens of all ages by generating new knowledge, developing innovative solutions and integrating where relevant a gender perspective to prevent, diagnose, monitor, treat and cure diseases. Horizon 2020 (H2020) was the EU’s multiannual funding programme between 2014 and 2020.
source link eu news
