Nordic Chapters Tech-Talk Series; Event 1: Acute Ischemic Stroke Analysis: Integrating Clinical Advancements and AI Solutions

#AIinStrorke #MedicalImagesandSignals #MeicalDataAnalysis #SaferStroke #WIE
Share

Tech-talk with title: Acute Ischemic Stroke Analysis: Integrating Clinical Advancements and AI Solutions

By: Associate Prof. Mahdieh Khanmohammadi 

Abstract:
Stroke remains a critical neurological emergency with significant societal and health impacts worldwide. In Norway, 15,000 people suffer from acute cerebral stroke annually, the leading cause of adult long-term severe disability, nursing home admissions, and the third leading cause of death. Rapid and accurate identification of ischemic regions in the brain is crucial for timely intervention and treatment planning in acute ischemic stroke patients. At Stavanger University Hospital, Computed Tomography Perfusion (CTP) scans and derived parametric maps are used to determine the need for immediate treatment with intravenous thrombolysis (IVT) alone or in combination with mechanical thrombectomy (MT), or MT alone. During follow-ups, patients undergo Diffusion Weighted Imaging (DWI) to evaluate the treatment outcomes.

 

This talk highlights a twin project combining clinical and technical research to enhance the diagnosis and treatment of acute ischemic stroke. The clinical PhD focused on optimizing stroke management through advanced imaging protocols and simulation-based team training, achieving significant reductions in treatment delays. The technical PhD explored artificial intelligence methods to automatically segment ischemic regions from CTP scans, emphasizing both irreversibly damaged core and salvageable penumbra. By leveraging supervised and semi-supervised machine learning approaches, the research demonstrated the feasibility of 4D CTP as a robust input for segmentation tasks. Now we are moving towards evaluating combinations of features in ischemic lesions at hospital admission CTP and follow-up DWI to predict outcomes and guide treatment, analyzing the effect of interventional treatments. Finally, we aim to implement the developed system in a clinical workflow for simulation training and research.



  Date and Time

  Location

  Hosts

  Registration



  • Date: 13 Feb 2025
  • Time: 01:00 PM to 02:30 PM
  • All times are (UTC+01:00) Copenhagen
  • Add_To_Calendar_icon Add Event to Calendar
If you are not a robot, please complete the ReCAPTCHA to display virtual attendance info.
  • University of Stavanger
  • Stavanger, Rogaland
  • Norway 4021
  • Building: Kjølv Egelands Hus
  • Room Number: KE E-439

  • Contact Event Host
  • Starts 27 January 2025 11:53 AM
  • Ends 13 February 2025 12:00 AM
  • All times are (UTC+01:00) Copenhagen
  • No Admission Charge


  Speakers

Mahdieh

Biography:

Mahdieh Khanmohammadi filed of research is medical signal/image processing, computer graphics, neural networks, and artificial intelligence. She finished her PhD degree at computer science department of university of Copenhagen (DIKU) in 2015 on behavioral stress analysis and modeling. She became an associate professor at the department of electrical engineering and computer science of university of Stavanger, Norway in 2019. Her focus is particularly in analysing medical signals and images to provide new insight into diseases such as acute stroke, Parkinson’s disease and cardiovascular disorders and insights that poses a computer modelling challenge.

Email:

Address:Norway





Agenda

 

13:00 - Event starts and audience gathering

13:15 - 13:50 Tech-talk by Associate Prof. Mahdieh Khanmohammadi from Dept. of Electrical Engineering and Computer Science at University of Stavanger; This part will be streamed via Teams.

13:50 - 14:00 Questions and comments.

14:00- 14:30 Discussion and Mingling, soft drinks and snacks will be available to the participants.