Transforming AI Integration in Healthcare
Over the past decade, artificial intelligence has emerged as a revolutionary technology with the potential to transform multiple sectors. However, the application of AI in healthcare, particularly its incorporation into medical devices, presents unique challenges due to the inherent complexity of the technology and the rigorous regulatory requirements.
The MedAIVeritas project, co-funded by Compete2030, proposes a multifaceted and collaborative approach towards the development of new validation processes for AI models in the medical field, combining the knowledge and expertise of different entities from the national scientific system and companies.
Context and Challenges
The growing use of AI in clinical environments aims to improve diagnosis, prognosis, and treatment personalisation. However, for AI models to be classified and used as medical devices, they must undergo rigorous regulation and validation processes that ensure not only their efficacy and safety, but also compliance in areas such as data protection and ethics.
Technological Complexity
Algorithm transparency and quality of available training data
Security and Protection
Personal data protection and cybersecurity against threats
Ethics and Trust
Ethical concerns, confidentiality, and user trust
Robust Validation Methodologies
The MedAIVeritas project's main objective is to transform the integration of AI in the healthcare sector, focusing on the development and evaluation of robust validation methodologies for AI models, in order to enable their integration into medical devices.
These methodologies enable the creation of new design, development, and validation processes, creating a simplified pathway for solutions to meet the necessary safety, ethics, regulation, and efficacy criteria.
The MedAIVeritas Consortium
The consortium is led by Complear and involves five distinct partners — CCG/ZGDV, ISEP, FFUP, FCUL, and Promptly. Each of these entities contributes specialised knowledge allocated to different tasks, from design and development to the validation and clinical certification of models.
Collaboration
Seven specialised entities working together
Innovation
Integrated scientific and business knowledge
Growth
Sustainable and scalable development
Validation
Rigorous clinical certification processes
Key Project Activities
Methodology Development and Selection
Clinical evaluation and validation of methodologies to ensure AI algorithm transparency
Security Strategies
Development of mechanisms to protect solutions against cyber threats and vulnerabilities
Ethical Guidelines
Creation of ethical frameworks that ensure responsible use of AI in medicine
Process Automation
Integration of results into digital products to automate AI model validation
Expected Results and Impact
The research results will lead to new AI model validation processes, which will be integrated into new digital products to automate AI validation in healthcare.
These processes will be integrated into Complear's digital compliance management platform and launched by Compliance Labs, establishing themselves as market standards and raising the barrier to entry in the sector.
The project aims to consolidate itself at the forefront of safety and efficacy in medical AI, promoting the recruitment of highly qualified human resources and strengthening innovation capabilities.
With a strong innovation component and strategic collaborations, MedAIVeritas will significantly contribute to the advancement of science and technology in the field of digital health.