Experimental assessment of infarct lesion growth in mice using time-resolved T2* MR image sequences

Cerebral ischemic stroke is a major reason for death in Germany and worldwide. Although lots of research has been done on longtime (3 hrs - 27 days) stroke growth, not much is known about the acute stroke growth. In this paper a method for the image based analysis and visualization of short-time infarct growth in murine model based on time resolved T2* MR image sequences is presented. After manual definition of initial healthy and infarct volumes the corresponding histograms are extracted and used for classification of the stroke tissues for every T2* image sequence. After automatic classification computed thresholds can be used to 3D visualize stroke growth dynamically and analyze the infarct size over time. The method proposed was validated by medical experts in terms of visual comparison to histological images of the brains after stroke. The medical experts reported that the accuracy of the projected infarct size is sufficient for further analysis steps of infarct growth. In summary the method proposed can help to understand the evolution of strokes and may help to get new insights in this disease and improve the therapy for stroke patients in future.