On x-ray videodensitometric blood flow measurement

Blood flow velocity is an important physiological parameter in studies of vascular systems and in diagnosis and treatment of vascular disease. The knowledge of instantaneous velocity of intracranial blood flow is important for planning, intra-operative monitoring and evaluating of endovascular treatment of neurovascular disease such as intracranial aneurysms, arteriovascular malformations and fistulae. Currently used approaches for precise instantaneous blood flow velocity measurements are inapplicable to intra-operative monitoring of endovascular treatment because of high invasiveness and low resolution.; We evaluate the efficacy of roentgen-videodensitometric blood flow and velocity measurement in a clinical neurointerventional environment. To that end, we reviewed the published videodensitometric algorithms, selected and implemented the ones best fit for the target environment. We also identified all major factors and parameters that may contribute to errors of roentgen-videodensitometric blood flow measurements, quantified their magnitudes and identified and implemented algorithms and strategies for reduction of influence of these factors on flow measurements. We evaluated the two selected videodensitometric algorithms using both simulated angiograms and angiograms obtained by imaging in vitro programmable flow phantoms in presence of clinically realistic distortions and other error-inducing factors. We also developed modifications of published algorithms to improve their performance in presence of clinically realistic error-inducing factors and evaluated them in the experimental environment.; Among them, the new Bolus Mass Tracking with Curve Fitting (BMT-F) method demonstrated the most accurate and reliable measurements on constant and slowly variant flows. The Distance-Density Curve Matching (DDCM) method was found to be highly vulnerable to contrast settling, although its accuracy and reliability has been reported to be good in presence of high contrast density gradients. No method produced reliable measurements when contrast density was constant or when the volume of contrast agent within the field of view was small.; The roentgen-videodensitometric approach to assessment of blood flow and velocity demonstrated a considerable promise as a tool for clinical use in a neurointerventional environment. Improvements in its accuracy and reliability can be obtained by optimizing contrast mixing and by using angiographic equipment capable of increased resolutions and of 3-D imaging.