Early Breast Cancer Detection Using 3D Ultrasound Tomography

Institution: University of California, San Diego
Investigator(s): Thomas  Nelson , M.D. -
Award Cycle: 2007 (Cycle 13) Grant #: 13NB-0176 Award: $225,000
Award Type: IDEA Competitive Renewal
Research Priorities
Imaging, Biomarkers, & Molecular Pathology>Developing and Improving imaging technologies: better and easier detection

Initial Award Abstract (2007)

Early detection is the most effective way to reduce deaths from breast disease, and mammography is the current “gold standard” for diagnosing breast disease. However, most cancers arise in dense ductal tissue, so lesion detection for women with dense breasts and at increased risk for breast disease is particularly challenging. Ultrasound is an important adjunct to mammography to identify, characterize and localize breast lesions, and it is not limited by dense breasts. Ultrasound also requires no radiation or compression. However, ultrasound is operator dependent and a lack of consistency between centers limits widespread acceptance. To be truly useful ultrasound must provide more consistent, high quality imaging with less operator dependence. This research proposes a new approach to ultrasound imaging wherein we will construct a breast ultrasound scanner to image the entire breast. This scanner design will provide consistent, high quality images of breast tissues especially in women with dense breasts or young women at increased risk who require more frequent imaging. The resulting high quality images should improve visualization of dense breasts and will improve early identification of breast cancer and help localize those cancers so they may be biopsied or treated more cost-effectively.

This IDEA competitive renewal project continues the work begun in our original CBCRP IDEA grant (2005) where we focused on designing and constructing a volume breast ultrasound (VBUS) scanner and performing an initial assessment of its performance in test objects and a limited number of volunteers. Scanner performance met the design objectives. In the next phase we will to continue adding refinements to improve scanner performance and expand our clinical studies to patients with breast tumors identified by mammography. We will compare scanner performance to breast CT, conventional ultrasound and mammography. We also will add blood flow imaging capability to the scanner and evaluate tumor vascularity (i.e., blood perfusion) in a group of patients with breast tumors. We plan to accumulate sufficient correlative imaging performance data to support an application for a larger clinical trial.

The development of a VBUS scanner should, (1) provide more consistent imaging and improve visualization of the architecture of dense breast glandular tissue, and (2) provide improved characterization of breast lesions and better localization of those lesions particularly in women with dense breasts or young women at increased risk who require more frequent imaging. Also, the VBUS scanner would be excellent for imaging contrast agents for blood flow and perfusion imaging. The lower cost of a VBUS scanner compared to full-field-digital-mammography (FFDM) and magnetic resonance imaging (MRI) also provides a more cost effective imaging modality available to a wider range of patients. Finally, since ultrasound is the modality of choice for biopsy guidance the improved data from a VBUS scanner would directly complement and benefit biopsy procedures.

Final Report (2011)

Breast cancer is a high-prevalence disease with early detection the most important factor determining survival. Detection in dense breasts is particularly challenging for mammography whereas ultrasound (US) performs superbly plus requires no radiation or compression. While US reflectivity and attenuation are well appreciated in clinical ultrasound imaging, sound speed also is important in differentiating tumors from cysts. Also, dense breasts have higher sound speed than fatty breasts and tumors are higher still. Our goal is to improve early detection of cancer in dense breasts. We have designed and built a Volume Breast Ultrasound (VBUS) imaging system that images the pendant breast, without compression. VBUS automates image acquisition, measures acoustic parameters (e.g. sound speed, attenuation, etc.) and improves image quality, overcoming conventional breast US shortcomings.

During this project we focused on four specific aims related to improving scanner performance, imaging patients with breast tumors, comparing imaging with multiple modalities, and evaluating tumor blood flow. In the first aim we focused on improving sound speed and attenuation image quality and made performance improvements to produce high-quality clinical images. We revisited basic acoustics and optimized reconstruction algorithms to improve sound speed and attenuation images. Scanner performance was validated using test objects simulating breast tissue that were imaged with mammography, CT, MRI and ultrasound. In the second and third aims we performed comparative studies between VBUS, mammography, contrast-enhanced breast MRI and contrast-enhanced breast CT that were encouraging. We struggled with limited access to patients with breast tumors resulting in fewer clinical studies than planned. In the fourth aim we developed specialized acquisition and signal processing algorithms for vascular imaging with encouraging initial results although without the anticipated FDA approval for US contrast studies imaged no patients.

Throughout the research we have significantly improved scanner performance and reproducibility for patient imaging. Sound speed and attenuation measurement show good results but will require more accurate determination of time delays and signal reduction. Vascular imaging showed good success but will not be possible in volunteers/patients until US agents receive FDA approval. The most challenging barrier was difficulty being able to image patients with breast cancer due to not having the scanner in the breast imaging clinic because of space limitations. Thus, while the VBUS scanner is fully functional for clinical scanning, accumulation of patients with breast cancer will take longer than the grant period. We are optimistic that we ultimately will successfully demonstrate the capability of VBUS scanning.

In this research we have designed, built and successfully tested a breast imaging system. VBUS images the pendant breast using off-the-shelf components resulting in an economical breast imaging system making this technology potentially available for widespread deployment in the breast imaging community. Patent disclosures have been filed related to VBUS design concepts. We have imaged volunteers and initiated an imaging protocol for young women with dense breasts in keeping with our emphasis on early detection in this underserved population where mammography does not perform well. Among our novel accomplishments is measuring sound speed and attenuation in the breast to produce images measuring additional tissue properties that could be useful for breast cancer diagnosis, using off-the-shelf components to significantly lower the breast scanner cost potentially increasing access to early detection. Finally, we are using VBUS volume imaging to provide guidance information to a robotic device to perform precision breast lesion biopsy promises to be an especially fruitful area for improving health care in the future.

Future research will focus on making continued improvements in determining the time delay for sound speed and attenuation measurement, including using iterative reconstruction methods to correct for refraction. We also will continue working with colleagues in the breast imaging center to expand access to patients with diagnosed breast cancer obtain correlative data to validate VBUS scanner performance. Initial results visualizing dense breast tissue architecture and glandular tissues in young women has been encouraging. Ultimately we will apply for grant funds for a larger clinical trial.

Symposium Abstract (2010)

Nelson TR, Nebeker J, Wallace AM , Ojeda-Fournier H

The goal of this project is improve biopsy accuracy and the biopsy experience for patients requiring a breast biopsy.  To do this we assessed the performance of dedicated volume breast ultrasound imaging (VBUS) system with compact robotic biopsy device to provide precision image-guided breast lesion biopsy. Improved detection and biopsy of small lesions is essential for improving breast cancer survival. Most cancers arise in dense ductal tissue where ultrasound performs better than mammography. Small lesions (~2-3 mm) are difficult to biopsy using hand held techniques. Minimally invasive robotic devices potentially can assist physicians perform more precise biopsies thereby improving breast diagnosis and management.

Our design has integrated our VBUS system with a compact robotic device having a 6-DOF articulated arm to reach any breast location within ±1.0 mm carrying up to a 3.0 kg load. A load sensor measured force (Fx, Fy, Fz), and torque (Tx, Ty, Tz) providing real-time data regarding biopsy device insertion and penetration forces. System performance was evaluated by scanning a variety of breast test objects having simulated lesions using a pendant patient breast position. We measured targeting error and reproducibility in air using acrylic spheres (3, 6, 9, 12, and 15 mm) and in GelWax based breast test objects with (lesion sizes 2-15 mm). Volume data provided 3-dimensional lesion coordinates. Targeting and guidance algorithms optimized the path for insertion of a Mammotome™ vacuum biopsy device. Physician guidance is used to direct robot motion and device insertion.

Overall the VBUS volume image data were acquired in 20 sec/slice (volume < 20 min.) showing ~1 mm spatial resolution with lesions clearly identified. Lesion targeting and guidance algorithms showed insertion trajectory prior to device motion. Initial positioning located the device adjacent to skin surface; insertion was under physician guidance. Robotic reproducibility was excellent with lesion targeting accuracy to within ±1 mm. Gel test objects provided force feedback data regarding object deformation to improve targeting small lesions.

Volume imaging improved lesion localization and biopsy device placement, especially for small lesions. Robotic devices may provide more precise device placement assisting physicians with biopsy procedures. This work demonstrates the potential to translate the capabilities of two rapidly developing areas of medicine: volumetric imaging and robotic devices into a fully-functional clinical volume image-guided, physician-directed robotic breast biopsy system.

Supported in part by CBCRP Grants: 11IB-0035, 13NB-0176, 15GB-0023