Click on image to view larger version.
Fig. 2H. —Scores for image quality were derived from four reviewers
using scale of 1–5 (1, an anatomic structure was well visualized in all
sections; 2, structure was well visualized in some sections; 3, satisfactory
visualization; 4, insufficient visualization; and 5, structure could not be
detected on image) to rate detectability of various anatomic structures. Two
to six typical organ features were described, depending on structure being
reviewed. Because of this procedure, these scores were not in original score
range of 1–5. Point clouds on graphs depicting scores for image quality
can be interpreted as follows: Medians of sum scores for image quality
produced by flat-panel detector radiography appear on x-axis.
Y-axis represents medians of sum scores for image quality produced by
computed radiography. Points on bisector of angle indicate that image quality
of two methods was rated as equivalent. Graphs on which points appear above
angle bisector indicate that image quality of digital flat-panel detector
radiography was rated better than that of computed radiography. Conversely,
graphs on which points appear beneath angle bisector indicate that image
quality of digital flat-panel detector radiography was judged to be poorer
than that of computed radiography. For depiction of pelvis, sacroiliac joint,
and head of femur, symmetric distribution of points above and beneath bisector
of angle indicates no significant difference between image quality of computed
radiography and digital flat-panel detector radiography.
