Erik L. Ridley, Kate Madden Yee, Abraham Kim, Wayne Forrest, and Brian Casey

Stop us if you've heard this before, but there was a dominant theme at this year's RSNA 2018 meeting in Chicago. And that theme, believe it or not, was artificial intelligence (AI). As at RSNA 2017, sessions on AI and the related topics of machine learning and deep learning dominated the halls of McCormick Place. However, the debate around AI has begun to shift in subtle ways.

December 12, 2018

2018 12 12 21 51 0657 Documenting Experience 400

2017

Top 5 trends from RSNA 2017 in Chicago

Sometimes it's hard to come up with a theme for a given RSNA conference, but in 2017 it's easy -- talk of artificial intelligence (AI) dominated the discussion at RSNA 2017. Attendees stuffed lectures on AI to the point where people were turned away from crowded rooms, while vendors took advantage of heightened interest to make claims about AI-powered products that sometimes strained credulity.

December 12, 2017

2017 12 12 22 22 1051 Top Trends Escalator 20171212221930

Page 1 of 1

Artificial Intelligence Ai Global Communication Network 400

Most FDA-cleared AI medical devices not tested on patient outcomes

Of the 1,059 FDA-cleared devices used in radiology, three had a registered trial.

Low-quality axial T2-weighted MRI scans with gradient-weighted class activation mapping. The low-quality MRI scans (top row) show blurred structures and artifacts making zonal and capsule boundaries indistinguishable. On the same scans with gradient-weighted class activation map overlays (bottom row), red indicates regions that were most heavily weighted by the deep learning model toward the low-quality prediction.

AI framework could improve MRI quality control in prostate cancer

(A) Ga-68 NK224 PET/CT maximum intensity projection image (left), axial PET/CT images (center), and axial CT images (right) in a 43-year-old female participant with lung adenocarcinoma show lesions in the right cerebellum (blue arrow), right frontal lobe (red arrow), and sternum (green arrow) with low, intermediate, and high tracer uptake, respectively. Photomicrographs (scale bar, 100 μm) with hematoxylin-eosin (H&E) staining (center bottom) and programmed death ligand 1 (PD-L1) immunohistochemical (IHC) staining (right bottom) show high PD-L1 expression in the sternal lesion (tumor proportion score ≥ 50%).

New immuno-PET tracer shows promise in lung cancer

$Artificial intelligence (AI) assistance facilitates correct fracture identification in cases initially missed by human readers. A representative case is shown using anteroposterior radiographs in a 29-year-old male patient with wrist trauma (suspected nondisplaced transverse scaphoid fracture). The reader’s initial unaided assessment (top) resulted in a missed fracture, whereas with AI assistance (bottom), the fracture was correctly identified. The green box represents the fracture location as predicted by the AI tool, the blue box represents the reference standard annotation, and the red dots represent individual reader annotations.$