Canon Medical Systems Corporation

Toshiba Medical Systems Corporation

Yujie Lu

Zhou Yu

Jian Zhou

Richard Thompson

Wenyuan Qi

Yi Qiang

Evren Asma

Liang Cai

Tzu-Cheng Lee

Jeffrey Kolthammer

Yu Zou

Xiaolan Wang

Manabu Teshigawara

Xiaohui Zhan

Qiulin Tang

Masakazu Matsuura

Karthikayan Balakrishnan

Ryo Okuda

Ting Xia

An information processing method of an embodiment is a processing method of information acquired by imaging performed by a medical image diagnostic apparatus, the information processing method includes the steps of: acquiring noise data by imaging a phantom using a medical imaging apparatus; based on first subject projection data acquired by the imaging performed by a medical image diagnostic modality of a same kind as the medical image diagnostic apparatus and the noise data, acquiring synthesized subject data in which noise based on the noise data is added to the first subject projection data; and acquiring a noise reduction processing model by machine learning using the synthesized subject data and second subject projection data acquired by the imaging performed by the medical image diagnostic modality.

Information processing method, medical image diagnostic apparatus, and information processing system

An apparatus, method, and computer-readable medium for improving image quality of a medical volume. In an embodiment, the apparatus includes processing circuitry configured to receive a reconstructed input image volume from X-ray projection data corresponding to a three-dimensional region of an object to be examined, apply a pseudo-three-dimensional neural network (P3DNN) to the reconstructed input image volume, the application of the pseudo-three-dimensional neural network including generating, for the reconstructed input image volume, a plurality of three-dimensional image datasets representing a different anatomical plane of the reconstructed input image volume, applying at least one convolutional filter to each of a sagittal plane dataset, a transverse plane dataset, and a coronal plane dataset, and concatenating results of the applied at least one convolutional filter to generate an intermediate output image volume, and generate, based on the application of the P3DNN, an output image volume corresponding to the three-dimensional region of the object.

Apparatus, method, and non-transitory computer-readable storage medium for improving image quality of a medical image volume

Devices, systems, and methods obtain first radiographic-image data reconstructed based on a set of projection data acquired in a radiographic scan; apply one or more trained machine-learning models to the set of projection data and the first radiographic-image data to obtain a set of parameters for a scatter kernel; input the set of parameters and the set of projection data into the scatter kernel to obtain scatter-distribution data; and perform scatter correction on the set of projection data using the scatter-distribution data, to obtain a set of corrected projection data.

Devices, systems, and methods for deep-learning kernel-based scatter estimation and correction

The present disclosure is related to removing scatter from a SPECT scan by utilizing a radiative transfer equation (RTE) method. An attenuation map and emission map are acquired for generating scatter sources maps and scatter on detectors using the RTE method. The estimated scatter on detectors can be removed to produce an image of a SPECT scan with less scatter. Both first-order and multiple-order scatter can be estimated and removed. Additionally, scatter caused by multiple tracers can be determined and removed.

Total SPECT scatter estimation and correction using radiative transfer equation

The present disclosure relates to an apparatus for estimating scatter in positron emission tomography, comprising processing circuitry configured to acquire an emission map and an attenuation map, each representing an initial image reconstruction of a positron emission tomography scan, calculate, using a radiative transfer equation (RTE) method, a scatter source map of a subject of the positron emission tomography scan based on the emission map and the attenuation map, estimate, using the RTE method and based on the emission map, the attenuation map, and the scatter source map, scatter, and perform an iterative image reconstruction of the positron emission tomography scan based on the estimated scatter and raw data from the positron emission tomography scan of the subject.

Method and apparatus for scatter estimation in positron emission tomography

A method and apparatus is provided to perform dead-time correction in a positron emission tomography (PET) by estimating a full singles spectrum using a scatter model. The scatter model can use a Monte Carlo method, a radiation transfer equation method, an artificial neural network, or an analytical expression. The scatter model simulates scatter based on an emission image/map and an attenuation image/map to estimate Compton scattering. In the full singles spectrum, the singles counts with energies less than 511 keV are determined from the simulated scatter. The attenuation image can be generated based on X-ray computed tomography or based on applying a joint-estimation to PET data.

Method and apparatus for singles spectrum estimation and for dead-time correction in positron emission tomography (PET)

X-ray scatter simulations to correct computed tomography (CT) data can be accelerated using a non-uniform discretization of the RTE, reducing the number of computations without sacrificing precision. For example, a coarser discretization can be used for higher-order/multiple-scatter flux, than for first-order-scatter flux. Similarly, precision is preserved when coarser angular resolution is used to simulate scatter within a patient, and finer angular resolution used for the scatter flux incident on detectors. Finer energy resolution is more beneficial at lower X-ray energies, and coarser spatial resolution can be applied to regions exhibiting less X-ray scatter (e.g., air and regions with low radiodensity). Further, predefined non-uniform discretization can be learned from scatter simulations on training data (e.g., a priori compressed grids learned from non-uniform grids generated by adaptive mesh methods).

Method and apparatus for fast scatter simulation and correction in computed tomography (CT)

A method and apparatus are provided for using a neural network to estimate scatter in X-ray projection images and then correct for the X-ray scatter. For example, the neural network is a three-dimensional convolutional neural network 3D-CNN to which are applied projection images, at a given view, for respective energy bins and/or material components. The projection images can be obtained by material decomposing spectral projection data, or by segmenting a reconstructed CT image into material-component images, which are then forward projected to generate energy-resolved material-component projections. The result generated by the 3D-CNN is an estimated scatter flux. To train the 3D-CNN, the target scatter flux in the training data can be simulated using a radiative transfer equation method.

Deep-learning-based scatter estimation and correction for X-ray projection data and computer tomography (CT)

A method and apparatus is provided that uses a deep learning (DL) network to improve the image quality of computed tomography (CT) images, which were reconstructed using an analytical reconstruction method. The DL network is trained to use physical-model information in addition to the analytical reconstructed images to generate the improved images. The physical-model information can be generated, e.g., by estimating a gradient of the objective function (or just the data-fidelity term) of a model-based iterative reconstruction (MBIR) method (e.g., by performing one or more iterations of the MBIR method). The MBIR method can incorporate physical models for X-ray scatter, detector resolution/noise/non-linearities, beam-hardening, attenuation, and/or the system geometry. The DL network can be trained using input data comprising images reconstructed using the analytical reconstruction method and target data comprising images reconstructed using the MBIR method.

Apparatus and method using physical model based deep learning (DL) to improve image quality in images that are reconstructed using computed tomography (CT)

A method and apparatus is provided to simulate and correct for scatter flux detected in a computed tomography (CT) scanner. The scatter flux from a bowtie filter and an anti-scatter grid are pre-calculated to generate respective scatter tables. Scatter from an imaged object is simulated for some views of a CT scan using a three-step radiative transfer equation (RTE) method. Using the simulated scatter flux from these views, an accelerated simulation method, such as a multiplicative method, an additive method, and a kernel-based method, can determine scatter flux for the remaining views. The spatial model for X-ray scatter from the object can be based on a reconstructed image of object, and can be segmented into organs and material components having different scatter cross-sections. A scatter model outside the imaging region can be extrapolated using low-dose scanning, a scout scan, and/or anatomical information.

Model-based scatter correction for computed tomography

A method and apparatus is provided to calculate scatter using a method to determine primary X-ray flux, first-scatter flux, and multiple-scatter flux using an integral formulation of a radiative transfer equation and using spherical-harmonic expansion. The integral for the primary X-ray flux does not include a spherical-harmonic expansion. The integral for the first-scatter flux includes an angle-dependent scatter cross-section. The integral for the multiple-scatter flux is performed iteratively, includes spherical harmonics, and includes a scatter cross-section expanded using Legendre polynomials. The integrals of attenuation along propagation rays can be accelerated using material decomposition of the attenuation coefficients. An anti-scatter-grid term can be included in the integrals to account for the effects of an anti-scatter grid on the fluxes prior to detection of the X-rays.

Growing community of inventors

Vernon Hills, IL, United States of America

Yujie Lu