Dear AuntMinnie Member,
Could ChatGPT pass the U.K.'s radiology fellowship exam? A research team from Birmingham, England, says it's possible, according to our top story of the week, finding that ChatGPT-4 answered 74.8% of true/false statements correctly.
Our second most-clicked story was posted in our Molecular Imaging content area, and it consisted of coverage of a study that found that FAPI radiotracer uptake on PET/CT imaging may be a biomarker for identifying early responders to rheumatoid arthritis treatment.
On a sobering note, our third most-read article of the week reported on the incidence of workplace violence, with researchers finding that more than 60% of diagnostic radiology and radiation therapy staff have experienced it.
Next up? Taking fourth place on the list was a study that suggested that arm position is a significant factor influencing radiation exposure to patients during whole-body PET/CT scans.
Our story on GE HealthCare's plans to acquire MIM Software garnered interest, with the company explaining that the purchase will help it further its effort to support AI-based segmentation and contouring and dosimetry analysis in the fields of radiology, molecular imaging, and radiation oncology. For more news like this, visit our Advanced Visualization content area.
Also this week, we highlighted research that shows AI's promise as a first reader for breast cancer screening and how multiparametric MRI can help predict breast cancer treatment outcomes; a study that suggested that ChatGPT-4 performs well in identifying incidental findings on CT via a process called single-shot learning; work that found nearly half of outpatient MRI orders are delayed; and a report on how the number of vacation days radiologists take may contribute to burnout.
Check out the complete list of the week's top stories:
- Could ChatGPT pass the U.K.’s radiology fellowship exam?
- FAPI-PET predicts treatment response in patients with rheumatoid arthritis
- Workplace violence ‘extremely high’ in radiologic sciences
- Patient position affects radiation exposure during PET/CT scans
- GE HealthCare to acquire MIM Software
- AI shows promise as first reader in breast cancer screening
- Multiparametric MRI helps predict breast cancer treatment outcomes
- ChatGPT identifies incidental CT findings
- Nearly half of outpatient MRI orders are delayed
- Vacation days taken and working during vacation tied to burnout
- Cannabis and the brain: What’s the long-term impact?
- High body fat raises fracture risk in women treated for breast cancer
- Point-of-care decision support reduces unnecessary CT, MRI
- VR test differentiates between novice, expert ultrasound users













![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.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)





