The chest x-ray has been a mainstay of radiology for more than a century, and several presentations at RSNA 2026 suggest its data may be more valuable than ever. Deep-learning models are finding new clinical signals in routine chest images and are turning the workhorse exam into an opportunistic screening tool for conditions well outside the lungs.
Attendees can expect to hear from researchers who have trained deep-learning algorithms to predict bone mineral density and detect osteoporosis directly from chest x-rays, including one group at Flinders University in South Australia working with more than 71,000 matched exams, and another testing a commercially available tool across 5,873 images from seven countries. Meanwhile, work from a team at Harvard University will show that deep learning applied to chest x-rays can predict cardiovascular risk on par with the American Heart Association’s standard PREVENT score, with no blood tests required.
Also on the AI side, researchers using the CheXthought dataset will share findings on how expert visual search strategies vary during chest x-ray interpretation, and what those patterns suggest for building AI that replicates expert cognition. In pediatric imaging, a group at Cincinnati Children’s Hospital addresses the dearth of algorithms in the specialty by presenting a deep-learning model that can segment and localize endotracheal tubes, tracheostomy tubes, and upper extremity catheters on children’s chest x-rays acquired in the ICU.
Speaking of the ICU, don’t miss a presentation on bedside efficiency, where researchers will offer data on an ultra-compact, AI-powered x-ray system roughly the size of a paperback book that cut total ICU imaging workflow time by more than three-quarters compared to conventional mobile equipment.
Among interventional radiology studies, researchers from Japan will address radiation safety, with data showing that standard lead protection shielded IR operators from DNA damage during TACE procedures. And a prospective registry study examines genicular artery embolization outcomes for knee osteoarthritis.
Also, two studies cast a spotlight on trends in real-world x-ray use. One tracks the sharp rise of e-scooter and e-bike trauma at a level I trauma center, and another shows that exams for 12 potentially low-yield anatomic sites have nearly quadrupled over 15 years.
Check out the presentations we’re highlighting below, and for even more on digital x-ray at the meeting, take a look at the RSNA 2026 program.
AI model segments lines and tubes on pediatric chest x-rays
Sunday, November 29 | 1:00 p.m.–2:00 p.m. | S4-SSPD01-8 | Room E352
A deep-learning model can segment and localize multiple lines, drains, and airway devices on pediatric chest x-rays, research in this session suggests.
Deep learning predicts bone density on chest x-rays
Sunday, November 29 | 1:00 p.m.-2:00 p.m. | S4-SSMK01-9 | Room E450A
An AI algorithm trained to predict bone mineral density from routine chest x-rays shows promise for opportunistic osteoporosis screening, according to this Sunday session.
Broader visual search tied to higher accuracy in chest x-ray reading
Monday, November 30 | 9:30 a.m.-10:30 a.m. | M3-SSIN02-2 | Room E450B
In this scientific session on imaging informatics, a study suggests that expert radiologists who scan more broadly when interpreting chest x-rays achieve meaningfully higher diagnostic accuracy than those with narrower patterns.
AI on chest x-rays matches PREVENT score for cardiac risk prediction
Monday, November 30 | 1:30 p.m.–2:30 p.m. | M6-SSCA03-1 | Room E352
In this session, a study will show that deep-learning models applied to routine chest x-rays can predict cardiovascular risk as accurately as the American Heart Association' standard PREVENT score
Chest x-ray AI detects osteoporosis across seven countries
Tuesday, December 1 | 1:30 p.m.-2:30 p.m. | T6-SSCH05-1 | Room E451A
An AI model that analyzes routine chest x-rays for signs of reduced bone mineral density held its own across a broad multinational dataset, according to research in this session.
Lead protection prevents DNA damage in TACE operators
Tuesday, December 1 | 1:30 p.m.-2:30 p.m. | T6-SSIR02-5 | Room E353B
Standard radiation safety protocols effectively shielded interventional radiologists from DNA damage during transcatheter arterial chemoembolization (TACE).
E-scooter, e-bike trauma drives rising imaging burden
Tuesday, December 1 | 3:00 p.m.-4:00 p.m. | T7-SSER01-2 | Room S406B
Encounters involving e-bike and e-scooter trauma at a level I trauma center climbed from a single case in 2017 to 111 in 2025, with a sharp rise in volume and its radiology-specific costs, according to this talk.
Ultra-compact AI X-ray unit slashes ICU bedside imaging time
Wednesday, December 2 | 9:30 a.m-10:30 a.m. | W3-SSER02-4 | Room E450A
An ultra-compact AI-powered x-ray system can cut bedside imaging workflow times in ICUs by more than 75% while preserving diagnostic performance, according to research in this session.
GAE delivers sustained pain relief in knee OA registry
Wednesday, December 2 | 3:00 p.m.–4:00 p.m. | W7-SSIR03-5 | Room E353A
Genicular artery embolization (GAE) reduced pain and improved function in patients with symptomatic knee osteoarthritis, according to this session.
Low-yield x-rays nearly quadrupled over 15 years at one center
Wednesday, December 2 | 3:00 p.m.-4:00 p.m. | W7-SSMK09-8 | Room E450B
In this musculoskeletal imaging session, a study will be presented showing that x-ray exam volumes for 12 anatomic sites considered potentially low-yield have nearly quadrupled over 15 years at one tertiary medical center.
