Introduction: Three-dimensional computed tomographic angiography (3-D-CTA) can be used to evaluate cerebrovascular disease. A potential limitation of this technology is obscuration of vascular anatomy by bone. In view of this, we developed a method for bone-free rendering using iterative relative fuzzy connectedness (IRFC) of 3-D-CTA to examine the cerebral vasculature without the intervening cranial base.
Methods: 3-D-CTA was obtained in 10 patients with cerebrovascular disease by use of an algorithm based on IRFC. Bone structures were removed and vascular anatomy was isolated with an almost completely automated process. Bone-removed images were rendered via maximum intensity projections, compared with digital subtraction angiography, and evaluated for vascular abnormalities.
Results: Compared with digital subtraction angiography, IRFC 3-D-CTA successfully defined the intracranial vascular anatomy in all 10 patients. An aneurysm was identified in 6 patients, bilateral carotid-cavernous fistulae in 1 patient, and no structural abnormalities in the remaining 3 patients.
Conclusion: In 3-D-CTA, our preliminary findings suggest that bone obscuration, one of the current limitations of CTA, can be overcome through appropriate computer algorithms. We are now working on improving the computational efficiency of these algorithms to allow routine use of IRFC 3-D-CTA in an interventional or surgical suite.