Method for separating arteries and veins in 3D MR angiographic images using correlation analysis

US6377835B1 · kind B1 · utility

Assignee

• Siemens Aktiengesellschaft · DE

Inventors

• Stefan O. Schoenberg · Kandel, DE
• Michael Bock · Pittsburgh, US
• Michael Knopp · Sandhausen, DE
• Gerhard Laub · Burlingame, US

Key dates

• Filing date: Aug 30, 2000
• Grant date: Apr 23, 2002
• Expiry date: Aug 30, 2020

Classification

• Technology area (CPC G): Physics
• CPC primary: G06T7/0012
• WIPO field: Measurement
• WIPO sector: Instruments

Abstract

In a 3D magnetic resonance angiography (MRA) method, a region of interest, containing a major pulmonary vessel, is defined in an examination subject, and the subject is injected with a contrast agent bolus and a number of 3D magnetic resonance angiography data sets are obtained from the examination subject before, during and after arrival of the bolus in the region of interest. A slice from each of the data sets is selected which contains the region of interest, and an average region of interest signal as a function of time is determined and stored as a reference time curve. In the respective selected slices, a signal-time curve is identified and each signal-time curve is cross-correlated with the reference time curve. The cross-correlation results are used to form a new three-dimensional data set containing arterial and venous correlation maps. Maximum intensity projections are computed from the arterial and venous correlation maps for producing a visualized image of the pulmonary vasculature in the region of interest, the arteries and veins in the respective maximum intensity projections being clearly distinguishable.