Automated Collateral Flow Assessment in Acute Ischemic Stroke Patients using Computed Tomography with Artificial Intelligence Algorithms
BACKGROUND Collateral circulation is associated with improved functional outcome in patients with large vessel occlusion acute ischemic stroke (AIS) who undergo reperfusion therapy. Assessment of collateral flow can be time consuming, subjective, and difficult due to complex neurovasculature. This study assessed the ability of multiple artificial intelligence algorithms in determining AIS patient collateral flow. METHODS 200 AIS patients between March 2019 and January 2020 were included in this retrospective study. Peak arterial computed tomography perfusion volumes were utilized to assess collateral scores. Neural networks were developed for dichotomized (≥50% or<50%) and multi-class (0% filling, 0%<filling<50%, 50%≤filling<100%, or 100% filling) collateral scoring. Maximum intensity projections from axial and anteroposterior (AP) views were synthesized for each bone subtracted three-dimensional volume and used as network inputs separately and together, along with three-dimensional data. 60:30:10 training:testing:validation splits and 20 iterations of Monte Carlo cross-validation were used. Network performance was assessed using 95% confidence intervals of accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). RESULTS The axial and AP input combination provided the most accurate results for dichotomized classification: accuracy=0.85±0.01, sensitivity=0.88±0.02, specificity=0.82±0.03, PPV=0.86±0.02, NPV=0.83±0.03. Similarly, the axial and AP input combination provided the best results for multi-class classification: accuracy=0.80±0.01, sensitivity=0.64±0.01, specificity=0.85±0.01, PPV=0.65±0.02, NPV=0.85±0.01. CONCLUSIONS This study demonstrated one of the first artificial intelligence based algorithms capable of accurately and efficiently assessing AIS patients' collateral flow. This automated method for determining collateral filling could streamline clinical workflow, reduce bias, and aid in clinical decision making for determining reperfusion eligible patients.
as reported in: World Neurosurg. 2021 Sep 7 [Epub ahead of print]