AAPM: Three modalities measure breast density, may predict cancer risk
Three different methods—cone-beam CT, breast MRI and dual-energy mammography— have proved accurate for measuring breast density, according to two post-mortem studies presented Sunday at the 52nd annual meeting of the American Association of Physicists in Medicine (AAPM) in Philadelphia.
Both studies were conducted by a group of medical physicists at the University of California, Irvine, led by Sabee Molloi, PhD, who runs the Imaging Physics Laboratory. The first study compared two existing techniques for measuring breast density—cone-beam CT and breast MRI, which found that both techniques gave highly similar estimates of the density of 20 pairs of breasts scanned post-mortem. The second study showed the promise of a third technique called dual-energy mammography for measuring breast density.
"A better measure of breast density should yield a more accurate assessment of risk for developing breast cancer," said Justin Ducote, a PhD candidate in the Imaging Physics Laboratory, who presented the study on dual-energy mammography. Measuring breast density is made difficult by the fact there is no currently accepted gold-standard method for doing so, he added.
In Ducote's study, the research applied dual-energy mammography to 20 pairs of postmortem breasts. The technique makes use of dual energy x-ray imaging, where overlapping tissue signals can be isolated and quantified by exploiting the change in x-ray attenuation at different energies. According to Ducote, this allowed breast density to be measured from digital mammograms.
Ducote and his colleagues found that the measurements of breast volume for the first (V1) and second (V2) orientation were related by V2 = 1.03 V1 – 6.2. The measurements of breast density for the first (D1) and second (D2) orientation were related by D2 = 0.97 D1 – 0.03. The correspondence of breast mass as calculated from dual-energy mammograms (MDE) to the mass of the sample as measured from a scale (MS) was related by MDE = 1.01 MS – 8.10.
Ducote's colleague Huy Le, PhD candidate, will present related research on Thursday. In Le's study, the group analyzed the ability of cone-beam CT and breast MRI to measure breast density in the same 20 pairs of postmortem breasts. They found that breast density measurements using these two techniques were highly correlated.
Le and colleagues found that breast weight calculated using MRI (W-MRI) and CT (W-CT) image data were related to the measured weight (WS) as W-MRI = 0.89WS + 50 and W-CT = 0.99WS- 14, respectively. Breast density computed from CT (D-CT) was related to densities computed from MRI (D-MRI) as D-MRI =0.96D-CT 0.0009. They concluded that breast density measurements using cone-beam CT and breast MRI were highly correlated.
"If we can get agreement of breast density measured on multiple imaging modalities, our confidence in the accuracy of the value we obtain will increase," said Le.
The next step, the researchers said, is to quantify the exact density of the breasts in the study through chemical composition analysis -- a destructive technique, which is why the research was done using postmortem tissue.
Both studies were conducted by a group of medical physicists at the University of California, Irvine, led by Sabee Molloi, PhD, who runs the Imaging Physics Laboratory. The first study compared two existing techniques for measuring breast density—cone-beam CT and breast MRI, which found that both techniques gave highly similar estimates of the density of 20 pairs of breasts scanned post-mortem. The second study showed the promise of a third technique called dual-energy mammography for measuring breast density.
"A better measure of breast density should yield a more accurate assessment of risk for developing breast cancer," said Justin Ducote, a PhD candidate in the Imaging Physics Laboratory, who presented the study on dual-energy mammography. Measuring breast density is made difficult by the fact there is no currently accepted gold-standard method for doing so, he added.
In Ducote's study, the research applied dual-energy mammography to 20 pairs of postmortem breasts. The technique makes use of dual energy x-ray imaging, where overlapping tissue signals can be isolated and quantified by exploiting the change in x-ray attenuation at different energies. According to Ducote, this allowed breast density to be measured from digital mammograms.
Ducote and his colleagues found that the measurements of breast volume for the first (V1) and second (V2) orientation were related by V2 = 1.03 V1 – 6.2. The measurements of breast density for the first (D1) and second (D2) orientation were related by D2 = 0.97 D1 – 0.03. The correspondence of breast mass as calculated from dual-energy mammograms (MDE) to the mass of the sample as measured from a scale (MS) was related by MDE = 1.01 MS – 8.10.
Ducote's colleague Huy Le, PhD candidate, will present related research on Thursday. In Le's study, the group analyzed the ability of cone-beam CT and breast MRI to measure breast density in the same 20 pairs of postmortem breasts. They found that breast density measurements using these two techniques were highly correlated.
Le and colleagues found that breast weight calculated using MRI (W-MRI) and CT (W-CT) image data were related to the measured weight (WS) as W-MRI = 0.89WS + 50 and W-CT = 0.99WS- 14, respectively. Breast density computed from CT (D-CT) was related to densities computed from MRI (D-MRI) as D-MRI =0.96D-CT 0.0009. They concluded that breast density measurements using cone-beam CT and breast MRI were highly correlated.
"If we can get agreement of breast density measured on multiple imaging modalities, our confidence in the accuracy of the value we obtain will increase," said Le.
The next step, the researchers said, is to quantify the exact density of the breasts in the study through chemical composition analysis -- a destructive technique, which is why the research was done using postmortem tissue.