(Left) The standard clinical PACS (Picture archiving and communication system) frontal chest X-ray image, which does not detect/identify LLIED. (Right) ZF GUI/Viewer display of 12-class AI model inference result on the same monitor. (IMAGE)

SPIE--International Society for Optics and Photonics

(Left) The standard clinical PACS (Picture archiving and communication system) frontal chest X-ray image, which does not detect/identify LLIED. (Right) ZF GUI/Viewer display of 12-class AI model inference result on the same monitor.

Caption

(Left) The standard clinical PACS (Picture archiving and communication system) frontal chest X-ray image, which does not detect/identify LLIED. (Right) ZF GUI/Viewer display of 12-class AI model inference result on the same monitor. The model immediately correctly detected and identified an MRI-unsafe LLIED from the same CXR. Radiologists can achieve positive adjudication with a single-click, confirmed by demarcation of the red boundary around the image. The AI model can, thus, deliver inference results to radiologists for adjudication before clinical use.

Credit

Credit: R. White et al., doi 10.1117/1.JMI.9.5.054504

Usage Restrictions

NONE

License

Original content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.