Advanced imaging needs protocols, say leaders.
Neuroradiology has a reputation for being at the cutting edge of imaging technology, and in many ways, these technologies have already made significant impacts on delivery-of-care and patient outcomes.
But according to neuroradiology leaders at this year’s American Society of Neuroradiology in Vancouver, advanced imaging might be a misnomer in some cases. While technologies can provide more superior outcomes, some of them can’t be considered “new” anymore. There are others, though, that are still gaining ground, and some that are still being tested.
Max Wintermark, MD, chief of neuroradiology at Stanford University, pointed to three technologies that demonstrate how an advanced imaging technique can move from being state-of-the art to common.
1. Diffusion weighted imaging: Used to identify acute ischemic stroke within the first few minutes of onset, this technology is also used to diagnose infection, abscesses, and brain tumors. It has migrated from being only used for head and neck to both spine and body imaging, as well. Its prolific use throughout MR, Wintermark says, means that while it is a superior technique, it’s now commonplace.
2. Perfusion imaging: This technique, used to examine how the brain receives its blood supply, is an effective way to identify which patients need stroke treatment. In fact, it’s now a recommended technique by the American Heart Association and the American Stroke Association. But, when it was first introduced in the 1980s, Wintermark says it was difficult to perform with CT and MR scanners, making it a less productive technology. Improvements in recent years, though, have moved perfusion imaging toward being a more widely used technique, and Wintermark suggests within five years it will be as standard as diffusion weighted imaging.
3. Diffusion Tensor Imaging (DTI): DTI determines how water travels along the white matter that connects different parts of the brain. But, it’s still very much in the investigational stage, Wintermark says. To date, it hasn’t been tested with a large enough population to determine a normal range for patients. More information must be gathered about a number of influencing factors, such as demographics, vascular risk factors, and severity of neurological and psychiatric conditions. Until then, DTI won’t be used as extensively even though it has the potential to become a mainstay advanced imaging technique.
The need for protocols
Just because advanced imaging techniques exist doesn’t mean everyone is using them the same way, says Alexander Norbash, MD, an interventional neuroradiologist from the University of California at San Diego. In many cases, there’s no standardized protocol for a vast majority of imaging sequences, so individual institutions perform them based on their own preferences and methods.
But, advanced imaging works best, he says, when facilities follow the same strategies.
“For example, 200 institutions could have 50 different ways of conducting pediatric pituitary imaging, and they can’t be compared because everything is in an unstructured fashion,” he says. “It’s an indistinct, chaotic, and illogical way to approach practice, but that’s what we currently have.”
To fix the problem, institutions, particularly academic medical centers, need to assume leadership and come together and discuss how to best use technologies to maximize imaging use.
“We have to sit together with other neuroradiologists, corral our wisdom, discuss our hypotheses and get feedback,” he says. “We must do this in a structured, cohesive way.”
Only, then, he says, can neuroradiology make effective comparative judgements about the best ways to use advanced imaging technologies.
Ultimately a facility’s innovators and managers need to collaborate more often.
“I suggest that managers responsible for maximizing efficiency spend time talking with innovators and innovators spend more time talking with managers,” he says. “Have conversations about your needs and work together. Reach across the aisle to work in an organized fashion, because that’s something we’re not doing very well right now.”
New Study Examines Short-Term Consistency of Large Language Models in Radiology
November 22nd 2024While GPT-4 demonstrated higher overall accuracy than other large language models in answering ACR Diagnostic in Training Exam multiple-choice questions, researchers noted an eight percent decrease in GPT-4’s accuracy rate from the first month to the third month of the study.
The Reading Room: Artificial Intelligence: What RSNA 2020 Offered, and What 2021 Could Bring
December 5th 2020Nina Kottler, M.D., chief medical officer of AI at Radiology Partners, discusses, during RSNA 2020, what new developments the annual meeting provided about these technologies, sessions to access, and what to expect in the coming year.