This work reports the use of ultrasound radio frequency (RF) time series analysis as a method for ultrasound-based classification of malignant breast lesions. The RF time series method is versatile and requires only a few seconds of raw ultrasound data with no need for additional instrumentation. Using the RF time series features, and a machine learning framework, we have generated malignancy maps, from the estimated cancer likelihood, for decision support in biopsy recommendation. These maps depict the likelihood of malignancy for regions of size 1 mm(2) within the suspicious lesions. We report an area under receiver operating characteristics curve of 0.86 (95% confidence interval [CI]: 0.84%-0.90%) using support vector machines and 0.81 (95% CI: 0.78-0.85) using Random Forests classification algorithms, on 22 subjects with leave-one-subject-out cross-validation. Changing the classification method yielded consistent results which indicates the robustness of this tissue typing method. The findings of this report suggest that ultrasound RF time series, along with the developed machine learning framework, can help in differentiating malignant from benign breast lesions, subsequently reducing the number of unnecessary biopsies after mammography screening.

An iterative solution to the inverse problem of elasticity and viscosity is proposed in this paper. A new dynamic finite element model that is consistent with known rheological models has been derived to account for the viscoelastic changes in soft tissue. The model assumes known lumped masses at the nodes, and comprises two vectors of elasticity and viscosity parameters that depend on the material elasticity and viscosity distribution, respectively. Using this deformation model and the observed dynamic data for harmonic excitation, the inverse problem is solved to reconstruct the viscosity and elasticity in the medium by using a Gauss-Newton-based approach. As in other inverse problems, previous knowledge of the parameters on the boundaries of the medium is necessary to assure uniqueness and convergence and to obtain an accurate map of the viscoelastic properties. The sensitivity of the solutions to noise, model and boundary conditions has been studied through numerical simulations. Experimental results are also presented. The viscosity and elasticity of a gelatin-based phantom with inclusion of known properties have been reconstructed and have been shown to be close to the values obtained using standard rheometry.

This paper introduces a new technique for the robust estimation of relaxation-time distribution in tissue. The main novelty is in the use of the phase of transfer functions calculated from a time series of strain measurements at multiple locations. Computer simulations with simulated measurement noise demonstrate the feasibility of the approach. An experimental apparatus and software were developed to confirm the simulations. The setup can be used both as a rheometer to characterize the overall mechanical properties of a material or as a vibro-elastography imaging device using an ultrasound system. The algorithms were tested on tissue mimicking phantoms specifically developed to exhibit contrast in elasticity and relaxation time. The phantoms were constructed using a combination of gelatin and a polyvinyl alcohol sponge to produce the desired viscoelastic properties. The tissue parameters were estimated and the elasticity and relaxation time of the materials have been used as complementary features to distinguish different materials. The estimation results are consistent with the rheometry, verifying that the relaxation time can be used as a complementary feature to elasticity to delineate the mechanical properties of the phantom.

The linear dynamic finite element model can be formulated such that the elasticity and viscosity of the elements appear as the parameters in a linear system of equations. The resulting system of equations can be solved directly using singular value decomposition or a similar technique or through defining a quadratic functional. A priori knowledge and regularity measures can be added as equality or inequality constraints. The sensitivity of the inverse problem solution to the displacement noise and model imperfections are tested in simulations, where the parameters were successfully reconstructed with a displacement signal-to-noise ratio as low as 20 dB. Also, the viscoelastic parameters have been successfully estimated for a breast phantom with an embedded hard inclusion. The study of the computation speed demonstrates the potential of the new method for real-time implementations.

The finite element method is commonly used to model tissue deformation in order to solve for unknown parameters in the inverse problem of viscoelasticity. Typically, a (regular-grid) structured mesh is used since the internal geometry of the domain to be identified is not known a priori. In this work, the generation of problem-specific meshes is studied and such meshes are shown to significantly improve inverse-problem elastic parameter reconstruction. Improved meshes are generated from axial strain images, which provide an approximation to the underlying structure, using an optimization-based mesh adaptation approach. Such strain-based adapted meshes fit the underlying geometry even at coarse mesh resolutions, therefore improving the effective resolution of the reconstruction at a given mesh size/complexity. Elasticity reconstructions are then performed iteratively using the reflective trust-region method for optimizing the fit between estimated and observed displacements. This approach is studied for Young's modulus reconstruction at various mesh resolutions through simulations, yielding 40%-72% decrease in root-mean-square reconstruction error and 4-52 times improvement in contrast-to-noise ratio in simulations of a numerical phantom with a circular inclusion. A noise study indicates that conventional structured meshes with no noise perform considerably worse than the proposed adapted meshes with noise levels up to 20% of the compression amplitude. A phantom study and preliminary in vivo results from a breast tumor case confirm the benefit of the proposed technique. Not only conventional axial strain images but also other elasticity approximations can be used to adapt meshes. This is demonstrated on images generated by combining axial strain and axial-shear strain, which enhances lateral image contrast in particular settings, consequently further improving mesh-adapted reconstructions.