Image processing for enhancement of ischemic stroke in computed tomography examinations

Stroke is one of the highest causes of death worldwide. Non-enhanced computed tomography (CT) and nuclear magnetic resonance imaging (MRI) are the two main imaging techniques used to detect stroke. CT has a lower cost and greater accessibility of the population, so it is still the main method used. In most cases, the assessment of the compromised brain area is performed subjectively and may lead to difficulties in diagnosis. This research proposes an approach based on a computational algorithm, highlighting regions of ischemic stroke. Different image processing methods were used to enhance ischemic tissues. A set of 41 retrospective CT scans from Botucatu Medical School (Brazil) was used, divided into 25 cases of acute ischemic stroke and 16 normal patients. Stroke cases were obtained within 4.5 h of symptom onset. After selection of CT slices, image averaging was performed to reduce the noise. This was followed by a variational decomposition model and the expectation maximization method was applied to generate enhanced images. We determined a test to evaluate the performance of observers in a clinical environment with and without the aid of enhanced images. The overall sensitivity of the observer’s analysis was 64.5 % and increased to 89.6 % and specificity was 83.3 % and increased to 91.7 %. These results show the importance of a computational tool to assist neuroradiology decisions, especially in critical situations such as the diagnosis of ischemic stroke.