Quantitative ultrasound augments conventional ultrasound information by providing parameters derived from scattering and attenuation properties of tissue. This presentation describes our work estimating attenuation (ATT) and backscatter coefficients (BSC), and computing effective scatterer sizes (ESD) to differentiate benign from malignant breast masses. Radio-frequency echo data are obtained from patients scheduled for biopsy of suspicious masses following an institutional IRB approved protocol. A Siemens S2000 equipped with a linear array and recently a volume scanner transducer is employed. Echo signal power spectra are computed from the tissue and from the same depth in a reference phantom having accurately measured acoustic properties. Ratios of the tissue-to-reference power spectra enable tissue ATT and BSC’s to be estimated. ESD’s are then computed by fitting BSC vs. frequency results to a size-dependent scattering model. A heterogeneity index HDI expresses variability of the ESD over the tumor area. In preliminary data from 35 patients, a Bayesian classifier incorporating ATT, ESD, and HDI successfully differentiated malignant masses from fibroadenomas. Future work focuses on analysis methods when diffuse scattering and stationary signal conditions, implicitly assumed in the power spectra calculations, are not present. This approach tests for signal coherence and generates new parameters that characterize these scattering conditions.