Papers Published

1. Sharma, A.C. and Soo, M.S. and Trahey, G.E. and Nightingale, K.R., Acoustic radiation force impulse imaging of in vivo breast masses, 2004 IEEE Ultrasonics Symposium (IEEE Cat. No.04CH37553), vol. Vol.1 (2004), pp. 728 - 31 .
   (last updated on 2007/04/13)

   Abstract:
   Acoustic radiation force impulse (ARFI) imaging utilizes brief, high energy, focused acoustic pulses to generate radiation force in tissue, and conventional ultrasonic correlation-based methods to track the resulting tissue displacements in order to image the relative mechanical properties of tissue. In an ongoing clinical study, ARFI datasets from in vivo breast masses are acquired prior to core biopsy. Matched B-mode and ARFI images are generated for each mass. Data sets are divided based upon biopsy results, and images are evaluated for differentiating features. The purpose of this study is to acquire in vivo ARFI datasets in real-time, and to identify differentiable features between benign and malignant breast masses in the ARFI images. A modified Siemens SONOLINE Antares™ scanner and a VF10-5 probe were programmed to implement ARFI imaging in a multi-focal zone configuration. Under an IRB approved protocol. patients scheduled for breast core biopsy were recruited for participation. Data was acquired in real-time and processed offline. Matched B-mode and ARFI images were evaluated concurrently. To date, 27 masses have been imaged under this experimental protocol. In addition, single focal zone ARFI data acquired from 52 masses with a Siemens SONOLINE Elegra™ scanner and a 75L40 transducer were evaluated for consistency of the differentiating features. Of the 27 masses interrogated via multi-focal-zone ARFI, 9 were malignant, 9 were benign fibroadenomas, 4 were cysts, and the rest were other benign masses (i.e. lymph nodes, fat necrosis, etc.) Structures in matched B-mode images are in good agreement with those in ARFI displacement images, with both modalities demonstrating comparable resolution. In general ARFI displacement images of malignant breast masses exhibit increased contrast and improved margin definition over matched B-mode images. Cancers displace less (i.e. they are stiffer) than the surrounding tissue, and generally appear larger than in matched B-mode images. In addition, some malignant masses exhibit a slower recovery time, which has not been observed with benign masses. The cysts and fibroadenomas, in general, exhibit less contrast in ARFI images than in matched B-mode images. In many cases, fibroadenomas are not clearly distinguished from the surrounding tissues in ARFI displacement images, and can appear either stiffer or softer than the surrounding tissue. Acoustic streaming is observed in cyst fluid in response to ARFI excitation. ARFI displacement images portray different, complementary information than matched B-mode images. Some possible differentiating features between malignant and benign breast masses have been identified by this pilot study. Promising features include: differences in B-mode and ARFI lesion size. displacement magnitude, recovery time, and image contrast. These results encourage further study of breast mass characterization using ARFI imaging

   Keywords:
   acoustic correlation;acoustic imaging;acoustic streaming;cancer;image matching;mammography;medical image processing;tumours;