Publication Cover Art on the cover of Annals of Biomedical Engineering June 2022 issue

The article 'Determining Clinically-Viable Biomarkers for Ischaemic Stroke Through a Mechanistic and Machine Learning Approach' has been selected for the cover art of the June edition of the Annals of Biomedical Engineering (2022).

Image used as the cover art for the Annals of Biomedical Engineering 2022 edition

Congratulations to Ivan Benemerito1,2, Ana Paula Narata3, Andrew Narracott1,4 and Alberto Marzo1,3 whose article 'Determining Clinically-Viable Biomarkers for Ischaemic Stroke Through a Mechanistic and Machine Learning Approach' has been selected for the cover art of the June edition of the Annals of Biomedical Engineering (2022).

  1. INSIGNEO Institute for In Silico Medicine, ºù«Ӱҵ, ºù«Ӱҵ, UK; 
  2. Department of Mechanical Engineering, ºù«Ӱҵ, ºù«Ӱҵ, UK; 
  3. Department of Neuroradiology, University Hospital of Southampton, Southampton, UK; 
  4. Department of Infection, Immunity and Cardiovascular Disease, ºù«Ӱҵ, ºù«Ӱҵ, UK

Ischaemic stroke (IS) occurs when a clot of blood obstructs a major artery and limits the blood perfusion to brain districts downstream the occlusion. The leptomeningeal anastomoses (LMAs) are small vessels that connect different parts of the brain and provide alternative perfusion pathways. Extensive LMA networks are associated with better post-stroke outcomes, but they can only be observed with imaging methods. Routinely used, non-invasive techniques such as transcranial Doppler ultrasound (TCD) can show whether the LMAs have been engaged or not but do not quantify the amount of distal perfusion, which is a quantity of clinical interest.

In this study we modelled the brain haemodynamics using a 1D approach and supported it with machine learning (statistical emulators) to perform a sensitivity analysis of the brain circulation. This enabled the identification of biomarkers related to the level of perfusion downstream the occlusion of the middle cerebral artery. These biomarkers, derived from quantities which are routinely measured in clinical scenarios, allowed the identification of threshold values for maximising the probability that observed values of the biomarkers correspond to good or poor overall distal perfusion. 

The identified biomarkers show potential for clinical translation and could potentially assist doctors in the preliminary evaluation of stroke patients. Working towards this, we have recently started a collaboration with the Imaging and Brain INSERM Research Unit at University Hospitals of Tours (France) to validate these biomarkers using mouse models and develop usltrasound-based methodologies for their measurement in a clinical setting. 

Acknowledgements:

European Union Horizon 2020 Programme: CompBioMed (Grant Agreement No 675451), CompBioMed2 (Grant Agreement No 823712)