A multi‐wavelength, laser‐based optical spectroscopy device for breast density and breast cancer risk pre‐screening

Abstract

Optical Breast Spectroscopy (OBS) has been shown to predict mammographic breast density, a strong breast cancer risk factor. OBS is a low-cost technique applicable at any age. OBS information may be useful for personalizing breast cancer screening programs based on risk to improve consensus on and adherence to screening guidelines. To facilitate the use of OBS in population-wide studies, a research prototype OBS device was modified to make it portable and cheaper and to require less operator interaction. Two major changes were made: (1) the broadband light source was replaced with a laser module with 13 individual wavelengths turned on sequentially, enabling the use of photodiode detectors instead of a spectrometer, and (2) the light sources and detectors were placed in fixed positions within 4 sizes of cup, eliminating the need for placement by the operator. Wavelengths were selected using data from two previous studies. The reduction in spectral content did not significantly reduce the ability to distinguish between different risk groups. Positions for the light sources and detectors were chosen based on Monte Carlo simulations to match the optically interrogated volumes of the original device. Two light sources and six detectors per cup were used in the final design.

Environmental exposures during windows of susceptibility for breast cancer: a framework for prevention research

“…In women under 40 years, two alternative imaging methods have been used to assess breast composition including three studies of magnetic resonance imaging (MRI) in women aged 15–30 years [188–190] and two of dual X-ray absorptiometry (DXA) in girls aged 10–16 years [191, 192]. In addition, optical spectroscopy (OS) provides a compositional view of the breast capturing variation in the amount of water, lipid, hemoglobin, and collagen, as well as overall cellular and connective tissue density [174–176]. Collagen density may promote epithelial cell proliferation and increase tumor mobility and invasion, while hemoglobin is associated with angiogenesis [193–195].…”

Optical assessment of mammographic breast density by a 12‐wavelength vs a continuous‐spectrum optical spectroscopy device

“…However, when the average PLS‐predicted MPD over all repeat PLS analyses is plotted against the LIBRA MPD, there appears to be a significant bias in the predicted PLS values for both the TiBS and the Cups spectra, with predicted MPD overestimated for lower MPD values and underestimated for values over approximately 20%. This could be due to boundary effects as the lower density breasts tend to scatter light less, leading to less optical confinement and a larger area sampled, which in turn could lead to a larger contamination of the signal by adjacent tissues such as the pectoral muscle .…”

“…The device, depicted in Figure , contains silicon photodiodes with a 33.6 mm 2 area and built‐in preamplifier (S8746‐01, Hamamatsu Corp, Hamamatsu, Japan). The light source wavelengths, source and detector positions and cup sizes selection has been described previously and was based on data from previous studies using the spectrometer‐based TiBS device . The source‐detector distances for each cup are listed in Table S1, Supporting Information, and an interior view of a cup showing the positions of the sources and detectors is shown in Figure S1.…”

“…A principal components (PCs) analysis was executed to reduce the dimensionality of the spectral data prior to logistic regression to predict high mammographic density. Principal components analysis (PCA) is a commonly used data analysis tool for spectral analysis and has been employed in past analyses of the TiBS device OBS data . PCA was performed separately on the spectra from the TiBS and the Cups devices, using the average distance‐corrected spectrum for each breast.…”

Breast density quantification using structured-light-based diffuse optical tomography simulations

“…This measured spectrum is analyzed using various methods to discriminate between BI-RADS I–III and BI-RADS IV categories [19] or used to predict the %BD [20]. Recent developments of this system include hardware portability and replacing the broadband source with 13 laser sources, while maintaining its ability to discriminate between low- and high-density cases [21]. Similarly, Taroni et al .…”