Нови изследвания ще използват изкуствен интелект за подобряване на резултатите при пациенти с инсулт

The EU-funded research project, VALIDATE, develops and validates an artificial intelligence (AI) based prognostic tool for predicting patient health outcomes in acute ischemic stroke. A decision support system will provide doctors and healthcare professionals additional information for choosing the optimal treatment to improve health outcomes and quality of life. 

Acute ischemic stroke affects more than one million people within the European Union every year. It occurs when a blood clot blocks a blood vessel, reducing the supply of oxygen to the brain. This can result in damage to brain cells and long-term disability for the stroke survivor such as speech impairment or paralysis.

The most effective treatment is mechanical thrombectomy (MT) where the blood clot is removed with a medical device to allow the blood supply back to the brain. If MT is not possible, the standard treatment is the application of a drug that can dissolve the blood clot, called intravenous thrombolysis. 

Both treatments (thrombectomy and thrombolysis) have been shown to be effective with favourable outcomes. However, the outcome differs for individual patients, as many different individual factors can influence the outcome of a stroke treatment such as co-morbidities, age, sex, genetic pre-disposition and lifestyle. VALIDATE will be refining and validating AI-based models to find out what combination of factors will influence patient outcomes. This integration of individual features of the patient will enable doctors to make the best stroke personalised treatment decision for each individual patient.

Potentially, an artificial Intelligence (AI)-powered Prognostic Tools and Clinical Decision Support Systems (CDSS) can make recommendations based on patient data at great speed, helping healthcare professionals in their decision-making process. CDSS’s have the potential for safer, faster, more accurate, and evidence-based individualised treatment plans, improved patient outcomes, and the cost-effective allocation of healthcare resources. 

Dietmar Frey, Charité Universitätsmedizin says, ‘By leveraging the available medical data and exploiting technological opportunities in the field of AI, and developing and validating trustworthy AI solutions to be implemented into clinical practice, we are seeking to surpass the clinical state-of-the-art by making a significant and sustainable impact on the treatment of acute stroke that will improve patient survival and outcome.’ 

Malte von Tottleben, Empirica adds, ‘The VALIDATE consortium combines excellence in technical and medical machine learning development with the clinical expertise of three leading stroke hospital partners. Additionally, our consortium benefits from the special expertise in the development of trustworthy AI, regulatory matters, ethics of AI, software design, and the translation of AI models to the clinical setting.’


Arlene Wilkie, Stroke Alliance for Europe says, ‘We are pleased to be involved in VALIDATE to ensure that the stroke survivor is involved in the development and testing of this new prognostic system, making sure it is fit for purpose for both the stroke survivor as well as the clinician. Ultimately, we hope that this new tool will a major impact on improving the outcomes for the millions of people who suffer an ischemic stroke throughout the world.’

VALIDATE hopes the results will form the basis of AI-based CDSS approaches in the clinical guidelines for acute stroke treatment.

Notes to editors

VALIDATE has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 777107.

More information can be found at https://validate-project.eu/


Contact details:

·       Dietmar Frey, Charité Universitätsmedizin Berlin Germany 


·       Malte von Tottleben, empirica Gesellschaft für Kommunikations- und Technologieforschung mbH Germany


·       Arlene Wilkie, Stroke Alliance for Europe



List of partners

·       Charité Universitätsmedizin Berlin Germany 

·       empirica Gesellschaft für Kommunikations- und Technologieforschung mbH Germany

·       Technological University Dublin Ireland


·       Universitätsklinikum Heidelberg Germany 



·       IBM iX - Aperto GmbH Germany 

·       SAFE Stroke Alliance for Europe Belgium 

·       NORA S.L. Spain


Regulatory boundaries:

Importantly, AI-based CDSSs will only be approved and accepted in the clinical setting if they fulfil all the necessary criteria with regards to regulatory, lawfulness, ethics, and robustness. The European Union has outlined trustworthy AI guidelines to address these criteria and these will be incorporated in the validation of AI-based CDSSs. A trustworthy AI approach will ensure that (systematic) automated and manual biases are taken into account, that patient subgroups will not be discriminated against, and that traceability, transparency, and auditability will be ensured during development, testing and validation.