AuntMinnie

Top eight incorrectly classified synthetic images (Most convincing fake images). Images A, B, C, D, E, F, G and H were the most infrequently correctly classified images at 24%, 43%, 59%, 64%, 69%, 71%, 71% and 73% respectively. All images are AI-generated. Images A, B and C are head CTs. Images D and E are chest CTs. Image F is a cardiac CT. Image G is a musculoskeletal MRI. Image H is a head MRI. Images are republished under a Creative Commons license (CC BY 4.0).

Radiologists can identify three out of four AI-generated images

Radiologists found it particularly difficult to distinguish between real and AI-generated CT and MR images.

Featured

Student Doctor Laptop Test 400 (1)

Petition seeks to increase number of attempts for taking ARRT exam

Money

2027 MPFS proposed rule would lift radiology reimbursement

Minnie Cast Thumbnail (15)

Women's Imaging MinnieCast, Episode 13: Partial breast irradiation in the spotlight

July 17, 2026

Annual and cumulative FDA authorizations of AI/ML-enabled medical devices, 1995–2025.

Radiology dominates thirty years of FDA AI device approvals

July 16, 2026

Societies applaud introduction of Patients First Act

July 16, 2026

Prostate Cancer 400

MRI could replace DRE for prostate cancer staging

July 15, 2026

Podcast: MRI tragedy, key aspects of the Colombini family's lawsuit

More episodes »

Check out our Cases of the Week!

60-year-old man with penile and scrotal swelling revisited

A 60-year-old man presented with penile and scrotal swelling.

32-year-old man with subacute left jaw pain revisited

40-year-old man undergoing x-ray after placement of pacemaker revisted

Minnie Cast Thumbnail (15)

Women's Imaging MinnieCast, Episode 13: Partial breast irradiation in the spotlight

MRI could replace DRE for prostate cancer staging

Petition asks FDA to implement new OEM rule for MRI safety

Incorrect AI suggestions influence reader performance on mammography

Podcast: MRI tragedy, key aspects of the Colombini family's lawsuit

Top eight incorrectly classified synthetic images (Most convincing fake images). Images A, B, C, D, E, F, G and H were the most infrequently correctly classified images at 24%, 43%, 59%, 64%, 69%, 71%, 71% and 73% respectively. All images are AI-generated. Images A, B and C are head CTs. Images D and E are chest CTs. Image F is a cardiac CT. Image G is a musculoskeletal MRI. Image H is a head MRI. Images are republished under a Creative Commons license (CC BY 4.0).

Radiologists can identify three out of four AI-generated images

Radiology dominates thirty years of FDA AI device approvals

Most patients want to know if AI being used for imaging reports

SIIM: CVMs on par with LLMs for labeling incidental breast findings

Women's Imaging MinnieCast, Episode 11: Could AI be the future of BI-RADS?

Ai 1

Provect AI wins FDA clearance for 3D imaging software

Varian highlights Fast-02 proton Flash therapy trial results

Raidium launches AI-native oncology imaging platform

GEHC, Catholic Health announce $500M imaging Care Alliance

Societies applaud introduction of Patients First Act

Maureen Kohi, MD

Pitt names new radiology department chair

Maureen Kohi, MD, joins the University of Pittsburgh after chairing the department of radiology at the University of North Carolina.

Petition seeks to increase number of attempts for taking ARRT exam

SDMS names new 2026 board, committee members

More Practice Management

Provect AI wins FDA clearance for 3D imaging software

By AuntMinnie.com staff writers

The software reconstructs volumetric 3D images from 2D C-arm x-ray data without requiring a CT scanner.

July 17, 2026

Varian highlights Fast-02 proton Flash therapy trial results

By AuntMinnie.com staff writers

Most participants reported complete pain relief at three months and with no adverse events from treatment.

July 16, 2026

(From left to right) Researchers Emily Daugherty, MD, Anthony Mascia, PhD, and John Perentesis, MD, inside the proton therapy treatment room at the Cincinnati Children's/University of Cincinnati Medical Center Proton Therapy Center.

Pitt names new radiology department chair

By AuntMinnie.com staff writers

Maureen Kohi, MD, joins the University of Pittsburgh after chairing the department of radiology at the University of North Carolina.

July 16, 2026

Maureen Kohi, MD

Raidium launches AI-native oncology imaging platform

By AuntMinnie.com staff writers

The company launched the platform at Moffitt Cancer Center in Tampa, Florida to replace a legacy radiomics tool.

July 16, 2026

GEHC, Catholic Health announce $500M imaging Care Alliance

By AuntMinnie.com staff writers

The strategic partnership covers covers equipment, service, and AI-enabled technologies across more than 40 Catholic Health sites.

July 16, 2026

SDMS names new 2026 board, committee members

By AuntMinnie.com staff writers

The new board directors and members of the finance and nominating committees will assume their roles this fall.

July 16, 2026

Vote Ballot Democracy 400 (1)

ASRT Museum and Archives launches self-guided video tour

By AuntMinnie.com staff writers

Museum visitors can explore the ASRT's history at their own pace.

July 16, 2026

Asrt Mus 84 Hdr

AdvanCell raises $315M for targeted alpha therapy

By AuntMinnie.com staff writers

The company is developing ADVC001, a lead-212 (Pb-212)-based targeted alpha therapy for metastatic prostate cancer.

July 15, 2026

Petition asks FDA to implement new OEM rule for MRI safety

Radiofrequency (RF)-induced burns prompt recommendations to MRI manufacturers.

July 14, 2026

Mri Scanner2 Social

Incorrect AI suggestions influence reader performance on mammography

By Amerigo Allegretto

The biggest impact on reader performance stemmed from false-negative AI suggestions.

July 14, 2026

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.

Podcast: MRI tragedy, key aspects of the Colombini family's lawsuit

Inside the Michael Colombini case, Gilk and Posh continue retrospective analysis.

July 14, 2026

The Invisible Force Podcast Album Cover Auntminnie Jan 2026 Thumbnails

Brain PET scans reveal neuron loss in long COVID

The findings suggest that treatments aimed at boosting dopamine release could be repurposed to treat long COVID.

July 14, 2026

Representative participant from each group shown. C-11 DTBZ non-displaceable binding potential (BPND) in voxels quantitated with simplified reference tissue model 2. Color scale indicates C-11 DTBZ BPND from 0 (black) to 3.5 (red), with reduced C-11 DTBZ BPND observed in the long COVID case compared to both healthy controls.