TY - JOUR
T1 - Computer-aided lesion diagnosis in automated 3-D breast ultrasound using coronal spiculation
AU - Tan, Tao
AU - Platel, Bram
AU - Huisman, Henkjan
AU - Sánchez, Clara I.
AU - Mus, Roel
AU - Karssemeijer, Nico
N1 - Funding Information:
Manuscript received November 07, 2011; revised January 10, 2012; accepted January 11, 2012. Date of publication January 16, 2012; date of current version May 02, 2012. This work was supported by the EU funded project HAMAM (IST-2007-224538) within the Seventh Framework Programme (FP7). Asterisk indicates corresponding author. *T. Tan is with the Department of Radiology, Radboud University Nijmegen Medical Centre, 6525 GA Nijmegen, The Netherlands (e-mail: [email protected]) H. Huisman, C. I. Sanchez, R. Mus, and N. Karssemeijer are with the Department of Radiology, Radboud University Nijmegen Medical Centre, 6525 GA Nijmegen, The Netherlands. B. Platel is with Fraunhofer MEVIS, 28359 Bremen, Germany.
PY - 2012
Y1 - 2012
N2 - A computer-aided diagnosis (CAD) system for the classification of lesions as malignant or benign in automated 3-D breast ultrasound (ABUS) images, is presented. Lesions are automatically segmented when a seed point is provided, using dynamic programming in combination with a spiral scanning technique. A novel aspect of ABUS imaging is the presence of spiculation patterns in coronal planes perpendicular to the transducer. Spiculation patterns are characteristic for malignant lesions. Therefore, we compute spiculation features and combine them with features related to echotexture, echogenicity, shape, posterior acoustic behavior and margins. Classification experiments were performed using a support vector machine classifier and evaluation was done with leave-one-patient-out cross-validation. Receiver operator characteristic (ROC) analysis was used to determine performance of the system on a dataset of 201 lesions. We found that spiculation was among the most discriminative features. Using all features, the area under the ROC curve (A z) was 0.93, which was significantly higher than the performance without spiculation features (A z = 0.90, p = 0.02). On a subset of 88 cases, classification performance of CAD (A z = 0.90) was comparable to the average performance of 10 readers (A z = 0.87).
AB - A computer-aided diagnosis (CAD) system for the classification of lesions as malignant or benign in automated 3-D breast ultrasound (ABUS) images, is presented. Lesions are automatically segmented when a seed point is provided, using dynamic programming in combination with a spiral scanning technique. A novel aspect of ABUS imaging is the presence of spiculation patterns in coronal planes perpendicular to the transducer. Spiculation patterns are characteristic for malignant lesions. Therefore, we compute spiculation features and combine them with features related to echotexture, echogenicity, shape, posterior acoustic behavior and margins. Classification experiments were performed using a support vector machine classifier and evaluation was done with leave-one-patient-out cross-validation. Receiver operator characteristic (ROC) analysis was used to determine performance of the system on a dataset of 201 lesions. We found that spiculation was among the most discriminative features. Using all features, the area under the ROC curve (A z) was 0.93, which was significantly higher than the performance without spiculation features (A z = 0.90, p = 0.02). On a subset of 88 cases, classification performance of CAD (A z = 0.90) was comparable to the average performance of 10 readers (A z = 0.87).
KW - Automated 3-D breast ultrasound
KW - Computer-aided diagnosis (CAD)
KW - Lesion segmentation
KW - Observer study
KW - Spiculation
UR - http://www.scopus.com/inward/record.url?scp=84860688845&partnerID=8YFLogxK
U2 - 10.1109/TMI.2012.2184549
DO - 10.1109/TMI.2012.2184549
M3 - Article
C2 - 22271831
AN - SCOPUS:84860688845
SN - 0278-0062
VL - 31
SP - 1034
EP - 1042
JO - IEEE Transactions on Medical Imaging
JF - IEEE Transactions on Medical Imaging
IS - 5
M1 - 6132427
ER -