Image guidance promotes safe and effective surgical management of a wide array of intracranial diseases. To better define the historical importance of image guidance and to assess the relative contribution of each imaging modality to the safety and efficacy of selected procedures, we reviewed our 20-year experience at a single institution. A retrospective review of our departmental surgical records was performed to identify patients who underwent brain surgery with image guidance between January 1979 and January 1999. We identified the use of intraoperative fluoroscopy , endoscopy, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, and angiography in 7,388 patients. During this 20-year interval, advances in neuroimaging were translated into the operating room environment. Fluoroscopic guidance received the highest overall rating and was deemed critical for the performance of successful transsphenoidal surgery (n = 436) and effective percutaneous trigeminal neuralgia management (n = 1,121). Ultrasound and angiography both had limited roles; the latter was important to successful outcomes in 64 patients undergoing aneurysm management (n = 64) and arteriovenous malformation Gamma Knife radiosurgery (n = 786). Endoscopy also had a small role but had limited cost. Beginning in 1982, a dedicated operating room CT scanner was used during both morphologic and functional stereotactic surgery (n = 1,749). After 1986, MRI was used increasingly in the management of selected functional and tumor cases (n = 337); despite great versatility for patients undergoing Gamma Knife radiosurgery, the costs were relatively high. Frameless neuronavigation (n = 263) had excellent versatility and was relatively low in cost. During the last 20 years, image guidance techniques have facilitated minimally invasive brain surgery at our institution. The relative merits of all these imaging tools depended mostly on their versatility and relative costs. Major centers currently contemplating the incorporation of image guidance into routine brain surgery need not reproduce our own learning curve.