Twisting Two into Three: What Problem Does It Solve?

Abstract

Three dimensions should be better than two for interpreting images. As coronary angiographers, we are used to looking at two-dimensional displays and interpreting images as three-dimensional structures, but we are often left wondering if our interpretation was right. In my opinion, our chief limitations, in order of frequency, are 1 difficulty laying out the views we need to assess, when a suspect area is hiding behind another vessel in every view tried, 2 limited cranial and caudal angles available for imaging (a problem that has not improved much with recent iterations of angiographic cameras and gantries), which creates obligatory foreshortening of some segments, 3 physics limitations with some patients, particularly the morbidly obese, which cause image resolution to fall off sharply, and 4 inattentiveness to a suspect area that gets missed, the embarrassing but not uncommon malady of the overly busy practitioner. When these pitfalls can be avoided, we do pretty well at integrating the two-dimensional image sets, recreating three-dimensional pictures in our brains and sorting out whether significant disease is or is not present. Our high procedural success rates and low complication rates speak to the consistency and effectiveness of our current approach.Still, we would like to be better at what we do, and we are sensitive to the limitations of angiography. IVUS has taught us unequivocally that we miss important lesions from time to time. Expect for item number 4, the list above consists of limitations imposed by the Address for reprints: Kirk Garratt, Fax: XXX; e-mail: kgarratt@ lenoxhill.net two-dimensional x-ray imaging systems we use. There is a sense that rendering angiographic images in three dimensions will help us avoid some of the pitfalls of a two-dimensional experience.For the most part, we have tried to enhance the three-dimensionality of our angiographic viewing experiences through trickery. Before leaving Mayo Clinic and moving east a few years ago, I was skilled at use of the patient cradle, a now nearly defunct tool for rotating a patient during angiography. With the cradle system, a patient could be positioned with more LAO or RAO angulation than could be achieved with the gantry alone, especially when the camera was also placed in a steep cranial or caudal position. In addition, the patient could be rolled much faster than most cameras could be rotated; so, quick adjustments on the fly were easy. However, patients hated the sensation of being tipped up on their sides and rolled around; and so, cradles have largely disappeared. Still, I found adding rotation to the image greatly helps the sense of three dimensions of the coronary tree.Some imaging companies have sought to improve the three dimensionality of angiograms by enhancing gantry system performance and, in addition, integrating imaging data to create more of a three-dimensional picture. This is possible largely because labs are going digital, and digital image processing has matured. For example, the Philips Integris line can employ a...