Finding such discrepancies is critical to the continuity of patient care, as medical records and reports are often utilized across multiple providers and facilities.
With the earth’s warmest March on record in 175 years barely in the rearview mirror, a team of radiology experts says medical imaging must step up its sustainability efforts.
Two advanced algorithms—one for CAC scores and another for segmenting cardiac chamber volumes—outperformed radiologists when assessing low-dose chest CT scans.
Given the subjective nature of diagnosing concussions, it can be difficult to make the call on whether a patient needs imaging, but many doctors err on the side of caution.
In a close collaboration with radiation oncologists, experts trained their model on the CT lung images of 787 patients and tested it on the scans of more than 1,300 patients from external datasets.
Nearly 20% of the women included in the study did not complete their follow-up imaging at all, and out of the 3,648 women who did adhere to additional imaging, 23.3% delayed their exams beyond the recommended 60-day period.
Though the participants’ actual ages ranged from 69 to 72 years old, the experts’ predictive analysis revealed their estimated current brain ages to be between 46 and 93.
Experts arrived at this conclusion after comparing incidents of hematoma between patients who did and did not receive antithrombotic therapy prior to their procedure.
Two advanced algorithms—one for CAC scores and another for segmenting cardiac chamber volumes—outperformed radiologists when assessing low-dose chest CT scans.
Raquel Roman, chair of the Radiology Business Management Association (RBMA) Young Professionals Committee, and director of growth at Essential Radiology, explains how the group mentors the next generation leaders.
![Advanced artificial intelligence (AI) models can evaluate cardiovascular risk in routine chest CT scans without contrast, according to new research published in Nature Communications.[1] In fact, the authors noted, the AI approach may be more effective at identifying issues than relying on guidance from radiologists. Representative non-contrast CT slices for two patients (left), with super-imposed segmentations (right). One artificial intelligence (AI) model was used to segment a cardiac mask.](/sites/default/files/styles/stack_views/public/2024-04/screen_shot_2024-04-23_at_10.44.32_am.png.webp?itok=vXt3hxDo)
