I usually believe to be a salutary exercise to reminisce our most salient personal mistakes.
When I was doing my neurosurgery residency in 1990, most of my colleagues looked pretty uncomfortable with electrophysiological diagnostic techniques, probably because, as opposed to imaging, these require a wider conceptual shift from the more concrete surgical experience. One of them asked me if I knew enough about ECG to help him out. Instead of doing just that, I gave him a lecture about how ECG was soon going to give way to vectorcardiography, a more mature, flexible and user-friendly technique. I tried to dissuade him from paying any more attention to ECG. Shame on me.
Although I had been interested in ECG as a medical student in the early 80s, I soon decided to give up any further study of it. “Why did they not use vectors in the first place?”, I asked myself. I decided that the most likely reason was the initial lack of hardware to make the calculations and images feasible. Since this fact was changing rapidly, I thought it would be a matter of just a couple of years until having multicolor rotating 3D vector graphs as the main initial diagnostic tool for heart disease. Vectors would make anatomical and functional interpretation of the electrical signals straightforward and amenable to further, more detailed analysis.
However, time went on. Many years after these events, I found myself studying ECG (again) in order to get my medical license in the US. While doing that, I laughed at my past naiveté. Of course, ECG is living and well; no clinician would ever think of switching it for anything different, and vectorcardiography is still at this moment considered “an experimental technique”, with no appeal to anybody but lab nerds. And my old neurosurgery colleague may still be angry at me.
The Brain Shop 