Fair enough. I can't fully explain how someone who had a traumatizing brain injury seen on an X-ray taken at one hospital and then a blessing was performed, the patient shipped to another hospital and all trace of the injury vanished on the next X-ray, but it happened. Just because it hasn't happened to you, doesn't mean faith-based healings or other miracles aren't real and repeatable. Maybe I don't have the theological training required to explain it, but I don't need to. It's something useful that happened to me and millions like me.
I'm sorry but you are wrong. What you described DOES NOT happen to millions. First off the vast majority of X-ray results give consistent results. If they didn't, they wouldn't be used to assist diagnosis. However, hospital's are not research labs, where experiments are verified and carefully examined before publication. They are production lines and like any production lines have a certain error rate. Misidentification of the actual X-ray, i.e. wrong patient, wrong X-ray examination, instrument malfunction, etc are likely causes.
Next miracles by definition do not happen everyday as a regular occurrence. Miracles are the exception that is used to justify the whole. Science uses the exceptions to refine or on occasion to offer a whole new explanation.
Can you imagine ANY religion taking an unexpected exception, e.g. a miracle and using that as a basis to say "Everything we have been saying for the last 100 years (1000 years) is wrong." And I don't mean a sect breaking off and forming a new religion.
Science does exactly that on a regular basis. Some theories have amassed so much evidence that they would take larger and larger exceptions to enable a new theory to supplant it, but is possible. When plate tectonics was proposed, it engendered a great deal of debate. But we didn't end up with Orthodox Geologists and New Revelations Geologists. The geologists who continued to try to prove the old theories, failed to do so. While the geologists using the theory of plate tectonics continued to make predictions and find more evidence that supported those predictions, i.e. delivered results.