Get more Radiology AI news with our free newsletter.

By signing up you content to our privacy policy

X-ray markers may skew AI’s ability to interpret extremity radiographs

The AUC for the original images with tech markers and anatomy was 0.84. Covering the markers yielded an increased AUC of .86 with heatmaps tending to redirect towards the bone, rather than the X-ray markers. When the anatomy was removed from images and the labels alone were evaluated, the AUC decreased to .64. Authors note that this indicates some labels could be more connected with anomalies than others. 

Since technologists are required to include image annotations to orient anatomy and positioning, researchers note that future development CNNs should take this into…

Read More

Radiology AI
Compare items
  • Total (0)