Dear AuntMinnieEurope Member,
Artificial intelligence is a red-hot topic within medical imaging right now, but what does it really involve and what effect will it have on clinical practice? Should radiologists be wary about it or embrace it with open arms?
Our informatics columnist, Dr. Neelam Dugar, has studied the evidence closely and attempted to answer these and other questions. Read more in the Imaging Informatics Community, or click here. And for further information, make sure you check out the new Artificial Intelligence Community on our sister site, AuntMinnie.com.
Anything that promises to improve communication with referring physicians is well worth a look. Researchers in Japan have reported the successful implementation of a PACS-based alert system for unexpected significant findings. They're convinced their system helps to avoid neglected reports and ensures that information can be more easily shared. For the full details, click here.
Germany's first Radiological Protection Act is likely to get the go-ahead in the summer and come into force next year. The subject is due to be discussed later this month at the German national congress in Leipzig. For an interview with Horst Lenzen from Muenster University Hospital, go to our CT Community, or click here.
Radiation dose also came under scrutiny last week at the International Conference on Nuclear Cardiology and Cardiac CT in Vienna. A U.S. group spelled out how a large nuclear cardiology lab was able to reduce its average radiation dose by 60% over the past eight years. Visit the Cardiac Imaging Community, or click here.
Preoperative planning still is mainly based on 2D datasets, which ignores the angle and the resulting distortion of the measured distances, according to German researchers. Furthermore, many procedures are planned on projection radiographic images, showing a significant distraction of the real anatomy. To tackle this problem, they've created virtual simulations using open-source computer-aided detection software. Learn more here.
![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnieeurope.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&dpr=2&fit=crop&h=167&q=70&w=250)
