To estimate local contraction from tagged MRI heartbeat images. and deduce the presence of non-contributing muscle (usually ischremic), it is necessary to map the heart contours accurately in each slice at each imaged time. Neither an algorithm nor a hwnan has been found able to do this on the evidence contained in a single slice; machine estimates must be corrected by hwnans, using criteria of 40 (space and time) consistency. This has taken a full working day per heartbeat, using only a 20 interface; in our Virtual Workbench environment for dextrous 'reach-in' work in virtual spaces, this time is reduced to less than an hour, making it practical to analyze larger numbers of cases.
CITATION STYLE
Serra, L., Poston, T., Hem, N., Ann, H. P., & Choon, C. B. (1995). Virtual space editing of tagged MRI heart data. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 905, pp. 70–76). Springer Verlag. https://doi.org/10.1007/978-3-540-49197-2_8
Mendeley helps you to discover research relevant for your work.