DETECTION OF CEREBRAL ISCHEMIA BASED ON MACHINE LEARNING

Project Information

DECIMAL

Grant agreement ID: 957004

DOI

10.3030/957004

Project closed

EC signature date 15 September 2020

Start date 1 October 2020

End date 30 September 2021

Funded under

• INDUSTRIAL LEADERSHIP - Innovation In SMEs

Total cost

No data

EU contribution

€ 137 747,50

Coordinated by

NICO-LAB BV
Netherlands

Periodic Reporting for period 1 - DECIMAL (DETECTION OF CEREBRAL ISCHEMIA BASED ON MACHINE LEARNING)

Reporting period: 2020-10-01 to 2021-09-30

Summary of the context and overall objectives of the project

Stroke is the second cause of morbidity and the leading cause of long-term disability. More than 1.1 million people in Europe suffer a stroke each year, which will increase to 1.5 million in 2025 due to an ageing population and unhealthy lifestyle. Stroke diagnosis and care is notoriously complicated. Some improvements have been made in the clinic, such as through the introduction of CT Perfusion imaging technology (CTP) to allow for quantitation. There are, however, many concerns with current implementations resulting in poor accuracy.

Nico.lab is an innovative start-up developing and marketing unique Artificial Intelligence (AI) technology which analyzes brain imagery – such as a CT or MRI scan – and provides health professionals with treatment advice. These analyses take just 3 minutes, thereby reducing time to diagnosis by hours and increasing objectivity. CTP is a vastly different technology and is thus not yet supported by our AI analysis. Therefore, we want to research and develop novel algorithms to natively analyze CTP data and provide quick treatment advice. We will thereby resolve shortcomings of the current technology and thus offer the best of both worlds: a clinically accepted, user-friendly protocol combined with our world-class accuracy and rapidity.

In this project we will hire an Innovation Associate that can combine medical, technical and scientific expertise to set-up R&D into novel CTP algorithms. The innovation associate will initiate and setup the research and development of CTP algorithms, specifically to analyze ischemic changes (blood flow) in the brain in people with stroke. With their specific knowledge they will analyze the technical and fundamental properties of CTP technology and translate this knowledge in a technical feasibility study in collaboration with the development team. This study will also result in a proof of principle algorithm. In parallel, they will research the practical feasibility of CTP integration into our StrokeViewer platform. The associate must align technical and practical feasibility, resulting in a comprehensive innovation roadmap for Nico.lab regarding CTP technology.

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

In this project NICO.LAB hired an Innovation Associate (IA) to kickstart the development of a novel CTP analysis algorithm. During this project, the IA has been responsible for mapping the CTP landscape in the Netherlands as well as the competitive field. The IA was heavily collaborating with NICO.LAB teams and several clinicians to develop an effective innovation strategy. This approach was highly successful; the CTP algorithm has been developed in a first effective prototype which is already in use at several hospitals. The IA has additionally helped develop a future development roadmap for NICO.LAB’s biomarker portfolio; it will soon become publicly clear in what ways the CTP algorithm will improve even further.

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)

The development of the CTP algorithm involved innovations on many fronts; technical, clinical and practical. From the start of the project our focus was on clinical relevance, traceability, accountability, high performance and trust. We have achieved all these goals. Our algorithm consists of multiple innovative aspects that analyze complex CTP scans in various manners. These analyses are then interpreted together to deliver a clinical decision support. This is especially important when dealing with CTP scans, as it has been shown that physicians are not well enough trained to analyze complex CTP scans manually. The performance of the algorithm outperforms competitors due to our unique approach. This will lead to quicker and more insightful assessment of stroke patients, which ultimately allows physicians to start treatment quicker, directly leading to better outcomes.

CTP screen