Abstract
Our aim was to compare human and computer accuracy in reading medical images of acute stroke patients. We analyzed data of patients who underwent assessment of Alberta Stroke Program Early CT Score (ASPECTS) and CT Perfusion (CTP) via Rapid Processing of Perfusion and Diffusion (RAPID) software RAPID ASPECTS, and RAPID CTP), compared to radiologist reports and manual measurements. We compared volumes calculated by RAPID CTP software with those selected by scanner-equipped software (GE). For reference, follow-up images were manually assessed in accordance with the Alberta Stroke Program Early CT Score (ASPECTS) territories retrospectively. Although exact ASPECTS score agreement between the automatic and manual methods, and between each method and follow-up, was poor, crossing of the threshold for reperfusion therapy was characterized by an 80% match. CT perfusion analyses yielded only slight agreement (kappa = 0.193) in the qualification of patients for therapy. Either automatic or manual scoring methods of non-contrast images imply similar clinical decisions in real-world circumstances. However, volume measurements performed by fully automatic and manually assisted systems are not comparable. Thresholds devised and validated for computer algorithms are not compatible with measurements performed manually using other software and should not be applied to setups other than those with which they were developed.
Highlights
Acute ischemic stroke is a leading cause of disability in the western world
In the advent of fibrinolysis, we learned the limitations of causative treatment strategies, only some of which have been revised in accordance with contemporary techniques of mechanical thrombectomy
We know that many perfusion-assisted stroke studies were conducted previously, employing numerous software and clinical setups, but the existence of evidence that the amount of viable ischemic brain tissue can be quantified is still arguable [5,6]. This implies another question: what constitutes the difference? Is it the novel—or superior but previously underestimated—algorithm for the assessment of cerebral blood flow parameters, or improvement in the remaining workflow components, such as filtering or segmentation? What is the share of importance in the final results provided by a perfusion examination? Does tissue function expressed by blood circulation give way to post-processing of the data acquired by CT scanners?
Summary
Acute ischemic stroke is a leading cause of disability in the western world. Considering the increasing proportion of elderly people in developed countries, we should expect that the associated socioeconomic burden will grow. When cornerstone articles on the widening of time windows were published [2,3], and most guidelines updated [4], a question was raised as to how strictly reperfusion qualification rules should follow those associated with the above-mentioned trials in order to replicate their results. We know that many perfusion-assisted stroke studies were conducted previously, employing numerous software and clinical setups, but the existence of evidence that the amount of viable ischemic brain tissue can be quantified is still arguable [5,6]. This implies another question: what constitutes the difference? This implies another question: what constitutes the difference? Is it the novel—or superior but previously underestimated—algorithm for the assessment of cerebral blood flow parameters, or improvement in the remaining workflow components, such as filtering or segmentation? What is the share of importance in the final results provided by a perfusion examination? Does tissue function expressed by blood circulation give way to post-processing of the data acquired by CT scanners?
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
