J. Correa et al., DIGITAL CHEST RADIOGRAPHY - COMPARISON OF UNPROCESSED AND PROCESSED IMAGES IN THE DETECTION OF SOLITARY PULMONARY NODULES, Radiology, 195(1), 1995, pp. 253-258
PURPOSE: To compare the accuracy with which simulated solitary pulmona
ry nodules can be identified on digital images of the chest that are u
nprocessed, processed with adaptive spatial filtering, or processed wi
th global filtering. MATERIALS AND METHODS: Six experienced chest radi
ologists evaluated 408 test radiographs (136 from each of the three ty
pes, half with simulated nodules) and judged whether a nodule was pres
ent. Data from the 2,448 observations were evaluated by means of a rec
eiver operating characteristic curve with location methods. RESULTS: A
ccuracy was significantly better with the adaptive filter technique th
an with the global technique (P < .05), and it was better with adaptiv
e filtering than with no processing in the detection of pulmonary nodu
les in the mediastinal-subdiaphragmatic areas (P < .05). No significan
t difference was found between no processing and global filtering. CON
CLUSION: Adaptive filtration is superior to global filtration in the i
dentification of solitary pulmonary nodules and is superior to no proc
essing in nodules projected over the radiopaque areas of the thorax on
digital images.