This year’s trip along the Road to RSNA for digital x-ray features a few familiar mileposts – AI for chest x-ray studies, for instance – but notably also significant research into how technology and new techniques can reduce radiation exposure in patients.
What’s changed in AI research is the number of studies evaluating commercially available chest x-ray algorithms in “real-world” settings. This is a testament to how research has shifted from standalone studies on the accuracy of algorithms to how they may actually impact clinical workflows – and thereby perhaps move a step closer to broader clinical implementation.
These studies include deployments of an AI fracture detection model in a hospital in Norway, an assessment of an algorithm in four sites in India, two sites in Korea, and one site in Thailand; a study of another algorithm deployed to identify suspected misplacements of endotracheal tubes; and a validation study of yet another model for autonomously reporting normal chest x-ray studies.
We’ve also highlighted below several sessions on new technology. In one, researchers discuss the benefits of outfitting portable x-ray machines with digital autoexposure control systems, while another highlights the clinical and environmental advantages of carbon nanotube sources for systems over standard filament tubes in fluoroscopic imaging – both of these new technologies can reduce radiation doses without sacrificing imaging quality, according to the presenters.
Another way to reduce radiation to patients would be to scale back on adjacent x-ray imaging, says a group in Australia, which will present a study on Sunday suggesting that a vast majority of second scans of the upper and lower limbs have no new findings.
Finally, we provide a sampling of interventional radiology research, which is also largely concerned with reducing patient radiation. One study suggests that digital variance angiography demonstrates a significant image quality advantage and enables radiation dose reduction in prostatic artery embolization (PAE).
Here, we’ve highlighted just a small sample of digital x-ray presentations scheduled for RSNA 2024. You can also view the complete 2024 scientific and educational program on the RSNA website.
DAEC system reduces portable x-ray radiation doses
Sunday, December 1 | 1:20 p.m.-1:30 p.m. | S4-SSPH02-3 | Room S404
In this session on physics in radiography, a study suggests portable x-ray systems outfitted with digital autoexposure control (DAEC) systems enhance image quality and reduce radiation doses. This latest technology holds promise, but has yet to be widely implemented in the field, the group noted.
Is adjacent x-ray imaging in the ER justified?
Sunday, December 1 | 2:30 p.m.-2:40 p.m. | S5-STCE1-1 | Learning Center Theater 1
A study in this session suggests that adjacent x-ray imaging had low diagnostic efficacy for both upper and lower limb regions. The finding underscores the importance of adopting a value-based approach to radiology, according to the researchers.
Are carbon nanotubes the future for digital x-ray systems?
Sunday, December 1 | 2:50 p.m.-3:00 p.m. | S5-STCE1-3 | Learning Center Theater 1
Carbon nanotube (CNT) sources for digital x-ray systems have clinical and environmental advantages over standard filament tubes in fluoroscopic imaging, according to a study presented in the session.
Digital variance angiography reduces radiation in PAE procedures
Monday, December 2 | 8:10 a.m.-8:20 a.m. | M1-SSIR01-2 | Room E353A
Digital variance angiography demonstrates a significant image quality advantage and enables radiation dose reduction in prostatic artery embolization (PAE), according to a study in this session. The finding is from a prospective trial and addresses critical safety considerations in PAE procedures.
Chest x-ray AI evaluated in clinical routine at a Norwegian hospital
Monday, December 2 | 1:30 p.m.-1:40 p.m. | M6-SSMK03-1 | Room E450A
In a musculoskeletal imaging scientific session, researchers at Vestre Viken Hospital Trust in Norway suggest that a fracture detection model improves clinical workflow, patient waiting times, and doctor consultations. The group notes that it is evaluating the model for implementation in four hospitals.
Chest x-ray AI performs well in postmarketing surveillance study
Tuesday, December 3 | 1:30 p.m.-1:40 p.m. | T6-SSCH06-1 | Room 451A
Even as the number of U.S.-cleared AI algorithms continues to increase, there's still a lack of data on how they perform in real-world clinical settings. In this session, an international postmarketing surveillance study will be presented that evaluated a commercial chest x-ray AI algorithm's performance in real-world clinical practice.
Study validates commercial AI for autonomous chest x-ray reporting
Tuesday, December 3 | 1:40 p.m. - 1:50 p.m. | T6-SSCH06-2 | Room E451A
Research conducted in the U.K. will be presented in this session that evaluated the performance of an AI algorithm designed to autonomously report high-confidence normal chest x-ray studies. With the current high burden of chest x-ray reporting on radiologists, such novel methods could be useful to streamline radiological assessments.
Is endovascular thrombectomy cost-effective?
Tuesday, December 3 | 3:50 p.m.-4:00 p.m. | T7-SSNPM02-6 | S402
During this session on noninterpretive skills, a study suggests that endovascular thrombectomy (EVT) is cost-effective in patients with basilar artery occlusion, from a U.S. healthcare perspective. The study comes on the heels of two recent studies showing the clinical benefit of the emerging procedure in patients up to 12 hours and between six and 24 hours from stroke onset.
Vision-language model exhibits bias reading chest x-rays
Wednesday, December 4 | 9:10 a.m.-9:20 a.m. |W2-STCE1-2 | Learning Center Theater 1
Clinical deployment of vision-language foundation models with inherent demographic biases could exacerbate healthcare disparities, especially for marginalized groups, according to a study to be delivered in this session.
AI identifies misplaced endotracheal tubes
Thursday, December 5 | 12:35 p.m.-12:45 p.m. | R5A-SPCH-7 | Learning Center
AI can significantly enhance patient safety by identifying suspected misplacements of endotracheal tube (ETT) placements, according to a presentation in this session. Moreover, the tool significantly reduced wait times for radiological assessment, according to the group.