Emerging technique contributes to fight against breast cancer

CHARLESTOWN, Mass., August 27, 2009 -- Researchers at the Martinos Center are working to improve the efficacy of breast cancer screening by combining X-ray mammography with the emerging technique Diffuse Optical Imaging (DOI). Using the two, complementary modalities, they hope to identify tumors sooner and with higher accuracy.

X-ray mammography has long been the standard for breast cancer screening, but the method still suffers a number of drawbacks. It yields a high rate of false-positives, resulting in many unnecessary biopsies. Imaging dense breasts, especially in younger women, has proved difficult. And the scans offer information primarily about tissue density; they reveal little about tissue functional status, including vascular and metabolic information, which could shed light on disease progression, for instance.

By measuring the concentrations of hemoglobin in tissue, Diffuse Optical Imaging can identify tumors sometimes even before they are structurally evident. Unfortunately, the images produced by the technology are generally low resolution, making it difficult to localize tumors within the breast.

Investigators in the Photon Migration Imaging Laboratory in the Martinos Center for Biomedical Imaging therefore have been working to combine the two modalities by developing a safe, inexpensive add-on to the traditional breast cancer screen diagnostic flow. The resulting system would offer all the strengths of the two techniques: namely, the high resolution and wide availability of mammography and the high image contrast and functional assessment of DOI.

They have established three milestones to evaluate the efficacy of this approach, said Qianqian Fang, an instructor in radiology at Harvard Medical School and assistant in biomedical engineering at Massachusetts General Hospital. The first is to image healthy breasts to determine the ranges of optically derived physiological parameters in the various tissues. Next, they will image and analyze the reconstructed results for breasts with malignant lesions to identify statistical differences from healthy tissue. The final milestone is to analyze breasts with benign lesions to identify statistical differences from both malignant lesions and healthy tissue, and thus to assess any improvement in specificity using the dual-modality system.

They reported the findings for the first milestone in an IEEE Transactions on Medical Imaging paper published earlier this year. After imaging 49 healthy subjects they established the bulk optical properties of healthy breasts and, importantly, determined that mammographic compression does not eliminate the optical image contrast. This had been a concern at the beginning of the study.

The next step is to identify the malignant tumor contrast and any differences between malignant and benign lesions. More than 100 subjects with breast lesions have been recruited and imaged since publication of the IEEE paper. The researchers are now developing the region-of-interest analysis and looking toward publication of results about imaging of breast lesions.

Contact: Qianqian Fang, Ph.D.