Stroke is a major cause of morbidity and mortality worldwide, mainly affecting the frail elderly population. Endovascular therapy with thrombectomy is highly effective, time sensitive treatment in patients with large vessel occlusion. This project aims to counter unwarranted variations in health services related to geographical distance and varying stroke expertise at primary stroke centres. If successful, this project may improve the quality and speed of work-flow in acute ischemic stroke care and increase the number of patients who receive early and highly beneficial endovascular treatment.
The main goal of this project is to increase the number of patients treated by thrombectomy and reduce inequality in health services through implementation of a tool to improve acute stroke diagnostics. We aim to organize and simplify the care pathway through a pragmatic approach to acute stroke imaging powered by cutting edge advances in image processing and artificial intelligence.
The present study is a regional, clinical study assessing the benefit of implementing an AI based automatic image analysis tool as a clinical decision support for radiologists on call. The study will start in the second half of 2022.
Novel AI-based image analysis tools applied to already available standard CT based imaging techniques can a) improve acute stroke diagnostics and b) increase the number of patients transferred and treated by thrombectomy.
- To assess the diagnostic accuracy of mCTA in detection of medium and large vessel occlusion ischemic stroke using AI-based analysis tools compared to assessment by radiologists on call.
- To assess if the implementation of available AI-based analysis tools applied to mCTA can increase the number of stroke patients eligible for and offered thrombectomy.
- To assess if the use of AI-based image analysis tools in radiological diagnostics in primary stroke centres can reduce the time from onset to recanalization in acute ischemic stroke patients treated with thrombectomy.
- To compare functional outcome and patient related outcome measures 3 months after thrombectomy in stroke patients who had their initial radiological diagnosis using AI-based image analysis tools to stroke patients diagnosed by standard care.
- To assess user experience related to feasibility, utility and satisfaction with the implementation of the AI-based image analysis tool.
Regional collaboration: This study is based on a broad multidisciplinary collaboration and interaction between primary and comprehensive stroke centres in North-Norway. Leaders and radiologists at Finnmarkssykehuset, Nordlandssykehuset and Helgelandssykehuset support the study and are involved in the planning and implementation of the project. The first phase of the project will include establishment of a regional network for acute stroke care with representatives from each hospital in the region. The infrastructure in collaborating institutions is ensured by enclosed statements by relevant leaders.
National collaboration: The project is coordinated with a similar project in Helse-Sør Øst, led by associate professor Anne Hege Aamodt at Oslo University Hospital (OUS). This project is assessing the use of StrokeSENS in a more densely populated environment with a potential for merged data analyses.
International collaboration: An international collaboration with leading neurointerventionists/ neurologists, Prof. Mayank Goyal and Prof. Michael Hill at University of Calgary, Alberta, Canada. They have been involved in the development of the StrokeSENS software and are currently running multicenter RCT trials on thrombectomy treatment of LVO and MVO. They will contribute with their extensive expertise on stroke and thrombectomy research and long experience in use of StrokeSENS for clinical image analysis.
The Trial Coordinating Centre will be based at UNN Tromsø. Principal investigator is Agnethe Eltoft. She will also be the main supervisor for the PhD student in this project. SKPI will participate in the validation of the AI software, and personell from SPKI will be involved in the image analysis. SPKI will also give advice on inquiries about technical and AI-related legal challenges, as well as guidelines for AI clinical research.