RSNA 2014 

Abstract Archives of the RSNA, 2014


BRS269

Clinical Validation of Combined X-ray and Optical Breast Imaging with 215 Lesion Cases

Scientific Posters

Presented on December 2, 2014
Presented as part of BRS-TUB: Breast Tuesday Poster Discussions

Participants

Qianqian Fang PhD, Abstract Co-Author: Research Grant, Koninklijke Philips NV
Bin Deng PhD, Presenter: Research Grant, Koninklijke Philips NV
Dana H. Brooks, Abstract Co-Author: Nothing to Disclose
Stefan Carp PhD, Abstract Co-Author: Nothing to Disclose
Richard H. Moore, Abstract Co-Author: Research support, General Electric Company Research support, Siemens AG Research support, Astrophysics Inc Research support, SAP AG
Daniel B. Kopans MD, Abstract Co-Author: Research Grant, General Electric Company Royalties, Cook Group Incorporated Consultant, Siemens AG
David Boas PhD, Abstract Co-Author: Research Grant, Koninklijke Philips NV Research Grant, Canon Inc

PURPOSE

Combining the structural information from x-ray mammography with the functional information from tomographic optical breast imaging (TOBI) represents a promising direction towards further enhancement of imaging sensitivity and specificity for clinical breast cancer diagnosis. Over the past 8 years, we have conducted a clinical study, including 307 subjects with breast lesions and 118 healthy volunteers, to validate the clinical efficacy of this new dual-modality approach.

METHOD AND MATERIALS

A total of 307 subjects with breast lesions were recruited between 2006 and 2013 under a HIPAA compliant protocol approved by the IRB. Among them, 215 subjects were found to have valid reconstructions and x-ray discernible lesions, including 99 malignant, 83 solid benign and 33 cystic lesions. All recruited subjects were scanned by a combined TOBI and tomosynthesis (DBT) system developed by our group. To utilize the spatially co-registered imaging information, we have developed a joint x-ray/TOBI reconstruction algorithm. Paired and two-sample t-tests are then used to find significant differences between different tissue types in the same breast as well as between lesion groups.

RESULTS

The reconstructed optical images with the structural-priors show more spatial details in the lesion region compared to those from the non-prior-guided reconstructions. One-tailed paired t-tests reveal significance differences in scattering coefficients (μs) at 830 nm between all pairs in adipose, fibroglandular tissues and lesions for all lesion groups. Significant difference in total hemoglobin concentration (HbT) between fibroglandular tissue and malignant tumors was also found (p=0.0009). Two sampled t-tests demonstrate significant differences in HbT (p=0.0139) and μs (p=0.0108) between malignant and solid benign lesions. The μs difference between solid benign and cystic lesions is nearly significant (p=0.0509).

CONCLUSION

The improved image quality achieved by fusing structural information in the optical reconstructions and the enhanced statistical power provided by a large study population has provided further evidence to support a combined x-ray/TOBI system as a clinically feasible approach for assisting breast cancer diagnosis.

CLINICAL RELEVANCE/APPLICATION

Combining optical imaging with x-ray mammography brings valuable functional assessment to the clinical evaluation of breast cancer, thus, leading to reduced unnecessary biopisies and missed cancers.

Cite This Abstract

Fang, Q, Deng, B, Brooks, D, Carp, S, Moore, R, Kopans, D, Boas, D, Clinical Validation of Combined X-ray and Optical Breast Imaging with 215 Lesion Cases.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045795.html