AIRSPACE: Advanced Imaging for Risk stratification of Stroke PAtients using Ct Examinations


Ischemic stroke is a dominant cause of death and disability in the Western world. A highly effective therapy that can completely reverse the symptoms is available by immediate administration of intravenous thrombolysis (IVT). Unfortunately, not more than 10% of patients qualify for IVT. One important reason for this is that IVT can cause substantial adverse effects including hemorrhage.  IVT is only allowed within the first 3-4.5 hours after stroke-onset, when tissue damage and the subsequent risk of hemorrhagic complication are assumed to be limited [1]. Alternative treatment options including intra-arterial thrombolysis and thrombectomy are also associated with substantial risk. Multimodal CT imaging has been proposed to support (extended) treatment decisions. Two large clinical trials (DUST, Utrecht and MrClean, Amsterdam and Rotterdam and Amsterdam) currently investigate the prognostic value of CT angiography (CTA) and CT perfusion (CTP) scans in acute stroke. These studies investigate the presence of collateral circulation (CTA), infarct core and penumbra and permeability of the blood brain barrier (BBB) as potentially important factors in risk stratification and therapy decision. The development and evaluation of novel image processing methods to accurately and objectively derive these and new quantitative imaging biomarkers from CT data is required, in order to study their potential for improved treatment selection in the acute stroke setting.

To this end, the STW CARISMA project AIRSPACE will advance individual CT-based diagnosis in stroke by (i) automatic extraction of quantitative data including thrombus size and location and estimation of degree of collateral flow from CTA, (ii) highly sensitive BBB permeability estimation by the introduction and validation of more accurate methods such as bidirectional kinetic models, (iii) introduction of hemodynamic vessel flow information taken from CTP and (iv) clustering of image data to achieve thin slice, high resolution perfusion images with unaltered acquisition protocol. Finally to facilitate integrated CT evaluation all existing and newly developed CTA and CTP feature maps will be combined to aid fast clinical decision making. A prototype that integrates all components of the project will be implemented using the existing Philips EBW clinical workflow. As the EBW is already used by the multicenter trials partners, all data of the world’s largest stroke image data bases will be available in AIRSPACE for development and validation.


Project Team:

  • Dr. ir. Hugo de Jong: - project-leader
  • Dr. Henk Marquering:  - project-leader
  • Prof. Dr. Wiro Niessen:  - project-leader
  • Edwin Bennink: - PhD Student  (started 1-9-2011)
  • Emilie Santos: AMC/ErasmusMC – PhD Student (started 1-5-2012)
  •  Vacancy – UMC Utrecht



Validation of CT brain perfusion methods using a realistic dynamic head phantom. Riordan AJ, Prokop M, Viergever MA, Dankbaar JW, Smit EJ, de Jong HW. Med Phys. 2011 Jun;38(6):3212-21.

Differences in CT Perfusion Summary Maps for Patients with Acute Ischemic Stroke Generated by Two Software Packages, F. Fahmi, H.A. Marquering, G.J. Streekstra, L.F.M. Beenen, B.K. Velthuis, E. VanBavel, C.B. Majoie, Geaccepteerd voor publicatie in AJNR


Two-Step Clustering Technique Can Halve Radiation Dose in CT Brain Perfusion Measurements, H W De Jong, A J Riordan, J Dankbaar, E J Smit, B K Velthuis, M Prokop, oral presentation at the 2010 RSNA meeting

Validation of CT brain perfusion analysis methods and protocols using a realistic dynamic hybrid head phantom, A J Riordan, J Dankbaar, E J Smit, M Prokop, H W De Jong, oral presentation at the 2010 RSNA meeting

Optimizing Temporal Resolution to Minimize CT Dose in CT Brain Perfusion, Riordan, A. J. Velthuis, B. Bennink, E. de Jong, H. W. , oral presentation at the 2012 ASNR meeting

Effects of Positive and Negative Shifts in Bolus Arrival Time on Deconvolution Methods for CT Brain Perfusion, Bennink, E. Riordan, A. J. de Jong, H. W. oral presentation at the 2012 ASNR meeting


Links: (under development)