The underlying theory of matched filtering (MF) in digital subtraction angiography (DSA) has been reviewed and examples of several applications have been given. MF represents a trade-off between SNR improvement and reduced temporal resolution. Perhaps the major limit of the MF technique is motion during the integration interval. Motion artifacts may come from two distinct sources: artifacts from moving noniodinated structures, and distortion from vessel motion during the integration interval. If extreme enough, motion will eventually corrupt all temporal subtraction schemes. However, these two artifacts may be addressed respectively by 'editing' an image sequence prior to input to the matched filter and by using EKG gating techniques. Instrumentation required to perform MF is not appreciably different from that for routine DSA. Enough storage and bits of precision are required to form linear combinations of a sequence of images.
|Original language||English (US)|
|Title of host publication||International Journal for Numerical Methods in Engineering|
|Number of pages||19|
|State||Published - Dec 1 1986|
ASJC Scopus subject areas
- Numerical Analysis
- Applied Mathematics