The following table or figure may be downloaded to PowerPoint for personal use in teaching and presentations. This feature is available to all subscribers to the journal.
You MUST read and follow the guidelines at Request to Reproduce AJR Content if you are distributing or using AJR content beyond academic use (limited distribution, non-revenue producing, or educational purposes).
(Downloading may take up to 30 seconds.
If the slide opens in your browser, select File -> Save As to save it.)
Click on image to view larger version.
Fig. 3A. —Graphs show correlation between computer output and average
radiologists' ratings without and with computer-aided diagnosis (CAD) scheme
for indicating likelihood of malignancy for lung nodules. 
= average
radiologists' ratings for malignant nodules, 
= average radiologists'
ratings for benign nodules, horizontal lines = range of radiologists' ratings
for each nodule. Graphs show correlation between computer outputs and average
radiologists' ratings without CAD (A) (r = 0.514) and with CAD
(B) (r = 0.784). Note that radiologists' ratings without CAD
scheme in some malignant (upper left quadrant) and benign (lower
right quadrant) nodules were obviously correct, whereas likelihood of
malignancy based on computer outputs alone was incorrect; even with incorrect
CAD outputs, radiologists retained correct ratings. One malignant case
(arrow) and one benign case (arrowhead) shown here are
illustrated in Figures 4A,
4B.
