RATIONALE AND OBJECTIVES: Develop a fully automated, objective method for evaluating morphology on breast magnetic resonance (MR) images and evaluate effectiveness of the new morphologic method for detecting breast cancers. MATERIALS AND METHODS: We present a new automated method (morphologic blooming) for identifying and classifying breast lesions on MR that measures margin sharpness, a characteristic related to blooming, defined as rapid enhancement, with a border that is initially sharp but becomes unsharp after 7 minutes. Independent training sets (98 biopsy-proven lesions) and testing sets (179 breasts, 127 patients, acquired at five institutions) were used. Morphologic blooming was evaluated as a stand-alone feature and as an adjunct to kinetics using free-response receiver operating characteristic and sensitivity analysis. Dependence of false-positive (FP) rates on acquisition times and pathologies of contralateral breasts were evaluated. RESULTS: Sensitivity of morphologic blooming was 80% with 2.46 FP per noncancerous breast: FPs did not vary significantly by acquisition times. FPs varied significantly by pathologies of contralateral breasts (cancerous contralateral: 4.29 FP/breast; noncancerous contralateral: 0.48 FP/breast; P < .0001). Evaluation of 45 cancers showed suspicious morphologies on 10/15 (67%) cancers with benign-like kinetics and suspicious kinetics on 5/10 (50%) cancers with benign-like morphologies. CONCLUSION: We present a new, fully automated method of identifying and classifying margin sharpness of breast lesions on MR that can be used to direct radiologists' attention to lesions with suspicious morphologies. Morphologic blooming may have important utility for assisting radiologists in identifying cancers with benign-like kinetics and discriminating normal tissues that exhibit cancer-like enhancement curves and for improving the performance of computer-aided detection systems.
RATIONALE AND OBJECTIVES: Develop a fully automated, objective method for evaluating morphology on breast magnetic resonance (MR) images and evaluate effectiveness of the new morphologic method for detecting breast cancers. MATERIALS AND METHODS: We present a new automated method (morphologic blooming) for identifying and classifying breast lesions on MR that measures margin sharpness, a characteristic related to blooming, defined as rapid enhancement, with a border that is initially sharp but becomes unsharp after 7 minutes. Independent training sets (98 biopsy-proven lesions) and testing sets (179 breasts, 127 patients, acquired at five institutions) were used. Morphologic blooming was evaluated as a stand-alone feature and as an adjunct to kinetics using free-response receiver operating characteristic and sensitivity analysis. Dependence of false-positive (FP) rates on acquisition times and pathologies of contralateral breasts were evaluated. RESULTS: Sensitivity of morphologic blooming was 80% with 2.46 FP per noncancerous breast: FPs did not vary significantly by acquisition times. FPs varied significantly by pathologies of contralateral breasts (cancerous contralateral: 4.29 FP/breast; noncancerous contralateral: 0.48 FP/breast; P < .0001). Evaluation of 45 cancers showed suspicious morphologies on 10/15 (67%) cancers with benign-like kinetics and suspicious kinetics on 5/10 (50%) cancers with benign-like morphologies. CONCLUSION: We present a new, fully automated method of identifying and classifying margin sharpness of breast lesions on MR that can be used to direct radiologists' attention to lesions with suspicious morphologies. Morphologic blooming may have important utility for assisting radiologists in identifying cancers with benign-like kinetics and discriminating normal tissues that exhibit cancer-like enhancement curves and for improving the performance of computer-aided detection systems.
Authors: Francesco Sardanelli; Gian M Giuseppetti; Pietro Panizza; Massimo Bazzocchi; Alfonso Fausto; Giovanni Simonetti; Vincenzo Lattanzio; Alessandro Del Maschio Journal: AJR Am J Roentgenol Date: 2004-10 Impact factor: 3.959
Authors: David A Bluemke; Constantine A Gatsonis; Mei Hsiu Chen; Gia A DeAngelis; Nanette DeBruhl; Steven Harms; Sylvia H Heywang-Köbrunner; Nola Hylton; Christiane K Kuhl; Constance Lehman; Etta D Pisano; Petrina Causer; Stuart J Schnitt; Stanley F Smazal; Carol B Stelling; Paul T Weatherall; Mitchell D Schnall Journal: JAMA Date: 2004-12-08 Impact factor: 56.272
Authors: Milica Medved; Xiaobing Fan; Hiroyuki Abe; Gillian M Newstead; Abbie M Wood; Akiko Shimauchi; Kirti Kulkarni; Marko K Ivancevic; Lorenzo L Pesce; Olufunmilayo I Olopade; Gregory S Karczmar Journal: Acad Radiol Date: 2011-10-01 Impact factor: 3.173
Authors: Alan I Penn; Milica Medved; Vandana Dialani; Etta D Pisano; Elodia B Cole; David Brousseau; Gregory S Karczmar; Guimin Gao; Barry D Reich; Hiroyuki Abe Journal: BMC Med Imaging Date: 2020-06-09 Impact factor: 1.930