ABSTRACT
A new method for a 3-D reconstruction of angiographic cerebral vessels was developed. This system was constructed from inexpensive equipment such as a personal computer, a video camera, a video display, and a digitizer. The input of the vessel images on anteroposterior and lateral angiograms was performed using a video camera with a personal computer and an image processing program (FDM 98-4; Fotoron Co.). These processed images were modified with our original editorial method. This can contribute a lot to the erasing of "non-vessel" images such as skull bone and other noise. It also adds concerning vessels which cannot be clearly taken from angiograms with the video camera. The most beneficial point of the editorial method was to reduce the time consumed in the reconstruction. The spatial points of X-ray discharging boxes against the AP and lateral films were calculated from the length and width of the radiated images of metal bars. These bars were attached to the 3 planes of the lucent box in which the patient's head was inserted to take the angiograms. The 3-D points of the vessels were calculated as follows; one point of a vessel on the lateral image and the spatial point of the X-ray discharging box against the image were linked by a line on which a point moved within a Z range. When the extended line which linked the point and the position of the X-ray discharging box against the AP view crossed the AP image, the Z value of the point was adopted and X and Y values were gained.(ABSTRACT TRUNCATED AT 250 WORDS)
