arXiv subjects

Thomas P. Matthews

Publications and source records attributed to Thomas P. Matthews.

6 recordsLinked to original sources

arXiv2020

A Multi-site Study of a Breast Density Deep Learning Model for Full-field Digital Mammography Images and Synthetic Mammography Images

Purpose: To develop a Breast Imaging Reporting and Data System (BI-RADS) breast density deep learning (DL) model in a multi-site setting for synthetic two-dimensional mammography (SM) images derived from digital breast tomosynthesis exams using full-field digital mammography (FFDM) images and limited SM data. Materials and Methods: A DL model was trained to predict BI-RADS breast density using FFDM images acquired from 2008 to 2017 (Site 1: 57492 patients, 187627 exams, 750752 images) for this retrospective study. The FFDM model was evaluated using SM datasets from two institutions (Site 1: 3842 patients, 3866 exams, 14472 images, acquired from 2016 to 2017; Site 2: 7557 patients, 16283 exams, 63973 images, 2015 to 2019). Each of the three datasets were then split into training, validation, and test datasets. Adaptation methods were investigated to improve performance on the SM datasets and the effect of dataset size on each adaptation method is considered. Statistical significance was assessed using confidence intervals (CI), estimated by bootstrapping. Results: Without adaptation, the model demonstrated substantial agreement with the original reporting radiologists for all three datasets (Site 1 FFDM: linearly-weighted $κ_w$ = 0.75 [95% CI: 0.74, 0.76]; Site 1 SM: $κ_w$ = 0.71 [95% CI: 0.64, 0.78]; Site 2 SM: $κ_w$ = 0.72 [95% CI: 0.70, 0.75]). With adaptation, performance improved for Site 2 (Site 1: $κ_w$ = 0.72 [95% CI: 0.66, 0.79], 0.71 vs 0.72, P = .80; Site 2: $κ_w$ = 0.79 [95% CI: 0.76, 0.81], 0.72 vs 0.79, P $<$ .001) using only 500 SM images from that site. Conclusion: A BI-RADS breast density DL model demonstrated strong performance on FFDM and SM images from two institutions without training on SM images and improved using few SM images.

eess.IV↗

arXiv2017

A forward-adjoint operator pair based on the elastic wave equation for use in transcranial photoacoustic tomography

Photoacoustic computed tomography (PACT) is an emerging imaging modality that exploits optical contrast and ultrasonic detection principles to form images of the photoacoustically induced initial pressure distribution within tissue. The PACT reconstruction problem corresponds to an inverse source problem in which the initial pressure distribution is recovered from measurements of the radiated wavefield. A major challenge in transcranial PACT brain imaging is compensation for aberrations in the measured data due to the presence of the skull. Ultrasonic waves undergo absorption, scattering and longitudinal-to-shear wave mode conversion as they propagate through the skull. To properly account for these effects, a wave-equation-based inversion method should be employed that can model the heterogeneous elastic properties of the skull. In this work, a forward model based on a finite-difference time-domain discretization of the three-dimensional elastic wave equation is established and a procedure for computing the corresponding adjoint of the forward operator is presented. Massively parallel implementations of these operators employing multiple graphics processing units (GPUs) are also developed. The developed numerical framework is validated and investigated in computer-simulation and experimental phantom studies whose designs are motivated by transcranial PACT applications.

physics.med-ph↗

arXiv2017

Mitigation of artifacts due to isolated acoustic heterogeneities in photoacoustic computed tomography using a variable data truncation-based reconstruction method

Photoacoustic computed tomography (PACT) is an emerging computed imaging modality that exploits optical contrast and ultrasonic detection principles to form images of the absorbed optical energy density within tissue. If the object possesses spatially variant acoustic properties that are unaccounted for by the reconstruction method, the estimated image can contain distortions. While reconstruction methods have recently been developed to compensate for this effect, they generally require the object's acoustic properties to be known a priori. To circumvent the need for detailed information regarding an object's acoustic properties, we previously proposed a half-time reconstruction method for PACT. A half-time reconstruction method estimates the PACT image from a data set that has been temporally truncated to exclude the data components that have been strongly aberrated. However, this method can be improved upon when the approximate sizes and locations of isolated heterogeneous structures, such as bones or gas pockets, are known. To address this, we investigate PACT reconstruction methods that are based on a variable data truncation (VDT) approach. The VDT approach represents a generalization of the half-time approach, in which the degree of temporal truncation for each measurement is determined by the distance between the corresponding ultrasonic transducer location and the nearest known bone or gas void location. Computer-simulated and experimental data are employed to demonstrate the effectiveness of the approach in mitigating artifacts due to acoustic heterogeneities.

physics.med-ph↗

arXiv2017

Deep Learning-Guided Image Reconstruction from Incomplete Data

An approach to incorporate deep learning within an iterative image reconstruction framework to reconstruct images from severely incomplete measurement data is presented. Specifically, we utilize a convolutional neural network (CNN) as a quasi-projection operator within a least squares minimization procedure. The CNN is trained to encode high level information about the class of images being imaged; this information is utilized to mitigate artifacts in intermediate images produced by use of an iterative method. The structure of the method was inspired by the proximal gradient descent method, where the proximal operator is replaced by a deep CNN and the gradient descent step is generalized by use of a linear reconstruction operator. It is demonstrated that this approach improves image quality for several cases of limited-view image reconstruction and that using a CNN in an iterative method increases performance compared to conventional image reconstruction approaches. We test our method on several limited-view image reconstruction problems. Qualitative and quantitative results demonstrate state-of-the-art performance.

cs.CV↗

arXiv2017

Regularized Dual Averaging Image Reconstruction for Full-Wave Ultrasound Computed Tomography

Ultrasound computed tomography (USCT) holds great promise for breast cancer screening. Waveform inversion-based image reconstruction methods account for higher order diffraction effects and can produce high-resolution USCT images, but are computationally demanding. Recently, a source encoding technique was combined with stochastic gradient descent to greatly reduce image reconstruction times. However, this method bundles the stochastic data fidelity term with the deterministic regularization term. This limitation can be overcome by replacing stochastic gradient descent (SGD) with a structured optimization method, such as the regularized dual averaging (RDA) method, that exploits knowledge of the composition of the cost function. In this work, the dual averaging method is combined with source encoding techniques to improve the effectiveness of regularization while maintaining the reduced reconstruction times afforded by source encoding. It is demonstrated that each iteration can be decomposed into a gradient descent step based on the data fidelity term and a proximal update step corresponding to the regularization term. Furthermore, the regularization term is never explicitly differentiated, allowing non-smooth regularization penalties to be naturally incorporated. The wave equation is solved by use of a time-domain method. The effectiveness of this approach is demonstrated through computer-simulation and experimental studies. The results suggest that the dual averaging method can produce images with less noise and comparable resolution to those obtained by use of stochastic gradient descent.

physics.med-ph↗

arXiv2015

A constrained variable projection reconstruction method for photoacoustic computed tomography without accurate knowledge of transducer responses

Photoacoustic computed tomography (PACT) is an emerging computed imaging modality that exploits optical contrast and ultrasonic detection principles to form images of the absorbed optical energy density within tissue. When the imaging system employs conventional piezoelectric ultrasonic transducers, the ideal photoacoustic (PA) signals are degraded by the transducers' acousto-electric impulse responses (EIRs) during the measurement process. If unaccounted for, this can degrade the accuracy of the reconstructed image. In principle, the effect of the EIRs on the measured PA signals can be ameliorated via deconvolution; images can be reconstructed subsequently by application of a reconstruction method that assumes an idealized EIR. Alternatively, the effect of the EIR can be incorporated into an imaging model and implicitly compensated for during reconstruction. In either case, the efficacy of the correction can be limited by errors in the assumed EIRs. In this work, a joint optimization approach to PACT image reconstruction is proposed for mitigating errors in reconstructed images that are caused by use of an inaccurate EIR. The method exploits the bi-linear nature of the imaging model and seeks to refine the measured EIR during the process of reconstructing the sought-after absorbed optical energy density. Computer-simulation and experimental studies are conducted to investigate the numerical properties of the method and demonstrate its value for mitigating image distortions and enhancing the visibility of fine structures.

physics.med-ph↗