Most of the scans I report are ordered for one reason: pain. Knee pain, back pain, hip pain, shoulder pain: The musculoskeletal radiologist’s working day is a long procession of suffering looking for an explanation. Sometimes I think my subspecialty could honestly be called the science of imaging pain.
Except that it can’t, and the reason it can’t is the most interesting thing about it.
I do not image pain. I image anatomy in people who happen to be in pain, hoping to find something that explains it. Pain itself has no imaging correlate in clinical practice. No sequence, however clever, shows it. To say it simply: Pain is invisible and personal; only the patient feels it. My job is to look at the images and hope to find its source.
And the translation fails constantly, in both directions. The patient in agony with a pristine scan is a weekly event. So is the “terrible” spine (desiccated discs, bulges at every level) in someone who came in for something else entirely and feels fine. If this were truly the science of imaging pain, it would be a failing science. The image and the experience disagree too often for comfort.
Real physician voices, twice a week
Free, and one click to unsubscribe.
But here is what struck me recently: I am not alone in this procedure. Consider the rheumatologist. A patient arrives with the same complaint, and the rheumatologist performs their own dissection, not of the anatomy but of the clinical picture: history, joint examination, bloodwork, the pattern of pain across the body and across time, hunting for the disease behind the symptom. Consider the psychotherapist. The patient brings psychological pain, and the therapist dissects a different interior: history, personality, patterns, the archaeology of a life, hoping to locate where the suffering comes from.
Three professions, one deep grammar: A suffering person presents, and the clinician searches a hidden interior for its origin. I dissect the anatomy, the rheumatologist the clinical picture, the therapist the life. Different territories, different instruments, the same hunt.
I should admit the parallel is not perfect, because the differences are instructive. My dissection is literal; theirs is metaphorical, and the word “dissect” shouldn’t be allowed to blur that. There is also a genuine asymmetry: For me, the search ends when the finding is made; I diagnose, and someone else treats. For the psychotherapist, the search is the treatment; the dissection doesn’t precede the therapy, it constitutes it. The rheumatologist stands between us: fluent in images like me, responsible for the person over time like the therapist. They are the ones who take my findings and do the very bridging this piece is about: deciding whether what was seen explains what is felt.
But all three of us share something less flattering than the hunt: its characteristic failure. The assumption that finding a source explains the pain. I find the disc bulge; the therapist finds the childhood wound, and either might be incidental. Every one of these professions is in the business of constructing satisfying origins for suffering, and the origin is sometimes real and sometimes an artifact of the search method itself. In my world, this is the logic behind over-imaging and the unnecessary surgery that follows: the compelling visible abnormality that must be the culprit, treated at the patient’s expense. Psychotherapy has its twin: the elegant story about the past that explains everything and changes nothing.
My own field shows how sharper tools can deepen the trap. As MRI resolution climbs, ultrasound keeps getting sharper and better, and AI learns to segment structures the human eye was never trained to see, there is a quiet race to describe new ligaments, new tendinous slips, new relationships between tendons and their neighbors, each one published, named, and added to the catalog of things a scan can find. The anterolateral ligament of the knee is the textbook case: described in 2013, debated ever since, and now routinely reconstructed, while its true contribution to knee stability is still argued over. But a newly visible structure is not a newly understood one. We rarely know which of these findings matter, which are normal variants, which are incidental passengers. Yet once a structure has a name and a picture, it becomes something to inject, to repair, to cut. Medicine has a habit of treating first and understanding later, and it is the patient who pays for the interval.
This is also where the machines stop. AI will soon segment every structure on every scan flawlessly; it can join the hunt better than any of us. What it cannot do is the thing the hunt most requires: telling a real origin from a satisfying one. That judgment (is this finding the cause, a contributor, or a bystander that happens to photograph well?) is not a pattern-recognition problem. It is a human one, exercised in the gap between what we can see and what the patient feels.
Bridging that gap is the clinician’s real work, and the bridge is experience itself. Finding the disc herniation or meniscal tear is pattern recognition; deciding whether the disc herniation or meniscal tear explains this patient’s pain is a different act altogether, built slowly out of thousands of cases: which findings mattered, which turned out to be bystanders, how the story usually goes. The radiologist does it by setting the image against the clinical picture; the rheumatologist by mapping the pattern across joints and testing it against bloodwork, time, and response to treatment; the psychotherapist by testing whether the elegant story about the past actually loosens the suffering in the present. The machine will soon find every abnormality on every scan; it has data, but it has no experience. It has never sat with a patient, never been wrong about a finding, never felt pain. And relevance is not in the image. It is in the encounter between a finding and a human experience the machine has never had.
So no, mine is not the science of imaging pain. It is something stranger and more honest: the disciplined search for pain’s sources, conducted in full knowledge that the search itself can manufacture what it finds. The most important skill in all three professions (radiology, rheumatology, psychotherapy) may not be finding the source. It may be knowing when to stop hunting.
Ramy M. Mansour is a musculoskeletal radiologist in Canada.


