Addressing alternate realities may be common practice for radiologists when they are subjected to questionable or negligible patient histories.
Long-term readers of this blog know I am more than a little fond of the sci-fi and fantasy genres, including superhero comics. It has been gratifying as outfits like Marvel have brought their offerings to screens big and small. I enjoy seeing the rest of society getting in on the good stuff … and feel bad for them when the occasional flop finds its way through whatever quality control should be going on with these endeavors.
In particular, it is nice to see some concepts of near-lifelong familiarity to me gracing the minds of other folks. For instance, if I have occasion to talk about alternate realities or the multiverse, I no longer have to go into a long-winded explanation that leads to peoples’ eyes glazing over, if not rolling back into their heads.
For most people, that may be the case. At least I can reasonably expect that the imminent release of the Doctor Strange in the Multiverse of Madness movie I referenced in the headline for this week’s blog has folks aware of the concept of a multiverse, right?
Okay, let’s assume not and cover all the bases. If you have seen stuff like Back to the Future, Star Trek, or Twilight Zone, you have been introduced to alternate realities. They can differ slightly or drastically from our “real” world. Plotlines usually showcase what would happen if a single key event happened differently (as pithily titled in Marvel’s “What if … ?” series). Sometimes it can be a more radical deviation, like when the Enterprise crew discovers another universe in which all of the heroes from this reality are villains in that one, and vice versa.
Gather up every single alternate reality you can think of (and recognize that there are infinitely more that you haven’t), and voila! The sum total is a multiverse. A vast collection of realities where anything that could happen did in infinite combinations with all of the other things that could/did happen.
It’s a fun premise for a writer because in an alternate reality, you can make up any storyline you like without consequences. It wasn’t the “real” timeline that future Marvel or Star Trek writers will have to work in and remain consistent with moving forward. You can kill off key characters, blow up planets, etc.
Again, this is far from a new concept for me. However, having a big-screen flick about to showcase it all probably has this on my mind more than it otherwise would be. While I was working this past weekend, I had the realization that I kind of practice radiology as if I were treating a multiverse’s worth of patients.
I have no idea what case I was reading when this notion hit. Odds are it was an ICU chest X-ray, just because they were the overwhelming majority of my day’s workload. Whatever it was, I was momentarily dwelling on the unreliable nature of the clinical histories, the “reasons for study” that auto-populate into my reports. For the gazillionth time, I mused that I have no way of knowing whether most of the histories I get are relevant or even accurate at all.
For instance, I would hazard a guess that at least 10 percent of the cases I receive are for “sepsis.” This is a very serious pathological entity. Even trying to treat one septic patient would be a huge burden, let alone 10 percent of your census.
The bright spot, however, is that most of the patients I receive imaging for in suspected “sepsis” cases do not actually have it. They might have an infection somewhere, and some mentally lazy or ignorant individual on the health-care team just writes “sepsis” because the clinician may think that is an acceptable usage of the word, and nobody will step in to correct him or her.
Maybe the patient is not infected at all. Perhaps it is an individual from a nursing home who can’t say what is ailing him, and the clinician ordering imaging is just thinking, “Could be sepsis” or “Let’s rule out sepsis.” Worse, it could just be some clerk who was told to get a computed tomography (CT) scan and rubber-stamped the requisition with “sepsis” because nobody would stand in the way of that.
Accordingly, when I get an imaging study for “sepsis,” I have to read it out as if a) the patient is truly septic, b) the patient isn’t frankly septic but might be infected somewhere, or c) there is no relevant clinical history whatsoever. There are surely more options I am not thinking of now.
I must address all of these possibilities. Not only do I want to do my best work for the patient, I also want to avoid a scenario in which some clinician calls me minutes, hours, or even days later to demand an addendum because now he or she now sees fit to provide me with the clinical history that I should have had in the first place. My report will cover the septic versions of this patient in one group of universes, the maybe infected but not septic versions in others, and the no relevant history versions in the remainder of the report.
Sepsis is far from the only nexus of alternate realities for a would-be multiversal radiologist. I routinely get x-rays for history of “fracture,” not knowing if that single word means the referrer knows there’s a fracture, thinks there’s a fracture, wonders if there might be a fracture, or wants to make sure there’s not a fracture.
I see the same thing for tube or line “placement.” More than occasionally when I get a case like that, whatever item the clinician referred to is not there. However, there is some other bit of support equipment that was not on the prior study. I have to read it out as if that new item is what the clinician was asking about but also as if the clinician placed a different item that has failed to reach the field of view (a central line that did not make it to the chest, for instance).
It’s like trying to diagnose Schrödinger's cat. I might never really know what the patient’s actual story was until after the fact. True, I could pick up the phone and try to track down whoever ordered the study but try doing that for each case you read in a day. I bet your RVU tally won’t be anything to brag about.
Can Radiomics Bolster Low-Dose CT Prognostic Assessment for High-Risk Lung Adenocarcinoma?
December 16th 2024A CT-based radiomic model offered over 10 percent higher specificity and positive predictive value for high-risk lung adenocarcinoma in comparison to a radiographic model, according to external validation testing in a recent study.
Can AI Facilitate Single-Phase CT Acquisition for COPD Diagnosis and Staging?
December 12th 2024The authors of a new study found that deep learning assessment of single-phase CT scans provides comparable within-one stage accuracies to multiphase CT for detecting and staging chronic obstructive pulmonary disease (COPD).