TY - JOUR
T1 - Incremental benefit of maximum-intensity-projection images on observer detection of small pulmonary nodules revealed by multidetector CT
AU - Gruden, James F.
AU - Ouanounou, Serge
AU - Tigges, Stefan
AU - Norris, Shannon D.
AU - Klausner, Todd S.
PY - 2002
Y1 - 2002
N2 - OBJECTIVE. Our purpose was to assess the incremental effect of maximum-intensity-projection (MIP) image processing on the ability of various observers to detect small (<1 cm in diameter) central and peripheral lung nodules revealed by multidetector CT. MATERIALS AND METHODS. We retrospectively identified 25 patients with metastatic disease, each having from two to nine nodules that were 3-9 mm in diameter. Two senior and three junior reviewers interpreted all images on a workstation. The observers first reviewed axial images (3.75-mm collimation, 3-mm reconstruction interval, multidetector acquisition) in cine and sequential fashion and recorded the size, lobe, and central or peripheral (within 1 cm of the edge of lung) location of each nodule. MIP images (10-mm slab, 8-mm interval) were then reviewed, and additional nodules detected were recorded. Final counts were established by consensus. RESULTS. The reviewers found 122 nodules (71 peripheral, 51 central) in the 25 patients. The addition of MIP slabs significantly enhanced reviewer detection of central nodules (p < 0.001) and junior reviewer detection of peripheral nodules (p < 0.001). MIP slabs also reduced the effects of reviewer experience, particularly for peripheral nodules. CONCLUSION. MIP processing reduces the number of overlooked small nodules, particularly in the central lung. Observer nodule detection remains imperfect even when lesions are clearly depicted on images.
AB - OBJECTIVE. Our purpose was to assess the incremental effect of maximum-intensity-projection (MIP) image processing on the ability of various observers to detect small (<1 cm in diameter) central and peripheral lung nodules revealed by multidetector CT. MATERIALS AND METHODS. We retrospectively identified 25 patients with metastatic disease, each having from two to nine nodules that were 3-9 mm in diameter. Two senior and three junior reviewers interpreted all images on a workstation. The observers first reviewed axial images (3.75-mm collimation, 3-mm reconstruction interval, multidetector acquisition) in cine and sequential fashion and recorded the size, lobe, and central or peripheral (within 1 cm of the edge of lung) location of each nodule. MIP images (10-mm slab, 8-mm interval) were then reviewed, and additional nodules detected were recorded. Final counts were established by consensus. RESULTS. The reviewers found 122 nodules (71 peripheral, 51 central) in the 25 patients. The addition of MIP slabs significantly enhanced reviewer detection of central nodules (p < 0.001) and junior reviewer detection of peripheral nodules (p < 0.001). MIP slabs also reduced the effects of reviewer experience, particularly for peripheral nodules. CONCLUSION. MIP processing reduces the number of overlooked small nodules, particularly in the central lung. Observer nodule detection remains imperfect even when lesions are clearly depicted on images.
UR - http://www.scopus.com/inward/record.url?scp=0036280811&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0036280811&partnerID=8YFLogxK
U2 - 10.2214/ajr.179.1.1790149
DO - 10.2214/ajr.179.1.1790149
M3 - Article
C2 - 12076925
AN - SCOPUS:0036280811
SN - 0361-803X
VL - 179
SP - 149
EP - 157
JO - The American journal of roentgenology and radium therapy
JF - The American journal of roentgenology and radium therapy
IS - 1
ER -