ACCURATE NONINVASIVE QUANTITATION OF BLOOD-FLOW, CROSS-SECTIONAL LUMEN VESSEL AREA AND WALL SHEAR-STRESS BY 3-DIMENSIONAL PARABOLOID MODELING OF MAGNETIC-RESONANCE-IMAGING VELOCITY DATA

Objectives. We present a new method in which a priori knowledge of the blood velocity fields within the boundary layer at the vessel wall, c ombined with acquisition of high resolution magnetic resonance imaging (MRI) blood velocity data, allow exact modeling at the subpixel level . Background. Methods are lacking for accurate, noninvasive estimation of blood flow, dynamic cross-sectional lumen vessel area and wall she ar stress. Methods. Using standard acquisition of MRI blood flow veloc ity data, we fitted all data points (n = 69) within the boundary layer of the velocity profile to a three dimensional paraboloid, which enab led calculation of absolute volume blood dow, circumferential vessel w all position, lumen vessel area and wall sheer stress. The method was tested in a 8.00 + 0.01-mm diameter glass tube model and applied in vi vo to the common carotid artery of seven volunteers. Results. In vitro the lumen area was assessed with a mean error of 0.6%. The 95% confid ence interval included the specified tube dimensions. Common carotid m ean blood flow was 7.42 ml/s, and mean (standard error) diastolic/syst olic vessel area was 33.25 (0.72 [2.2%])/43.36 (0.65 [1.5%]) mm(2). Me an/peak wall shear stress,vas 0.95 (0.04 [4.2%])/2.56 (0.08 [3.1%]) N/ m(2). Conclusions. We describe a new noninvasive method for highly acc urate estimation of blood flow, cross sectional lumen vessel area and wall shear stress. In vitro results and statistical analysis demonstra te the feasibility of the method, and the first in vivo results are co mparable to published data. (C) 1998 by the American College of Cardio logy.