General Electric Company

Ge Precision Healthcare LLC

Giovanni Palma

Razvan Iordache

Serge Muller

Ann-Katherine Carton

Zhijin Li

Sylvain Bernard

Serge Louis Wilfrid Muller

Laurence Vancamberg

Pablo Milioni De Carvalho

Ann-katherine Carton

Pablo Milioni DeCarvalho

The current disclosure provides for mapping medical images to style transferred medical images using deep neural networks, while maintaining clinical quality of the style transferred medical image, thereby enabling a clinician to evaluate medical images in a preferred style without loss of clinically relevant content. In one embodiment the current disclosure provides for a method comprising, acquiring a medical image of an anatomical region of a subject, wherein the medical image is in a first style, selecting a target style, wherein the target style is distinct from the first style, selecting a clinical quality metric, selecting a trained style transfer network based on the target style and the clinical quality metric, mapping the medical image to a style transferred medical image using the trained style transfer network, wherein the style transferred medical image is in the target style, and displaying the style transferred medical image via a display device.

Systems and methods for medical image style transfer using deep neural networks

An imaging system, such as a DBT system, is provided that is capable of providing processing of an image from initial generation/reconstruction of the image through the review of the image by a radiologist or other practitioner that produces images optimized for the particular task/process performed using the images. The task/review that the radiologist is performing on the processed or reconstructed image can be modeled with regard to the initial generation of the processed or reconstructed images from the raw images or the model considerations can be retroactively applied and back propagated through the image processing performed by the system. These modeling considerations that can take the form of an indicator of clinical performance is/are used by the system to optimize and produce an image(s) that best represents the items in the image necessary for an accurate diagnosis of the patient when reviewed by the radiologist.

Clinical task-based processing of images

A method of obtaining x-ray images includes controlling operation of an automatic injector to inject a contrast agent into a patient at a predetermined time, synchronizing operations of an x-ray imaging system with the operation of the automatic injector, and obtaining images of a region of interest of the patient during arterial and diffusion phases of the contrast agent.

Method and apparatus for contrast-enhanced spectral mammography programmed sequences

Abstract systems and methods of biopsy control include reconstructing a 3D volume from a plurality of tomosynthesis projection images and producing a plurality of synthetic stereo images from the plurality of tomosynthesis projection images. At least the synthetic stereo images are presented on a graphical display to a clinician to facilitate at least one input of a biopsy location for biopsy control.

Synthetic images for biopsy control

In the present invention, an image review method and system for performing the method is provided to facilitate the review of image volumes obtained in different imaging modalities and to evaluate the correlation between findings in the respective volumes in an easy to review manner compared with prior art systems and processes. The method and system of the invention provides simulated or actual combined or fused images or representations of registered images obtained in each imaging modality to simplify and accelerate the review of multi-modal breast imaging volumes, in particular when correlation of information between the two modes, including but not limited to, LE and DE, DBT and ABUS, DE and ABUS, etc., is important to evaluate.

Fused slice or cine-loop image for multi-mode DBT acquisitions

A method to generate a final image with a medical imaging system, wherein the final image highlights suspect regions in a tissue matrix and wherein the final image is generated from a first image and from previously acquired images of the tissue matrix is provided. The method comprises: creating a 3D map of suspicion from the previously acquired medical images; generating a 2D map of suspicion by projecting all or part of the 3D map of suspicion onto a plane from a point reproducing the conditions of acquisition of the first image; and generating a final 2D image of the tissue matrix from the first image and/or from the 2D map of suspicion, wherein the final 2D image highlights suspect regions in the tissue matrix.

Image processing method to determine suspect regions in a tissue matrix, and use thereof for 3D navigation through the tissue matrix

A derivation procedure for a three-dimensional digital mask from a series of two-dimensional masks in a radiographic device containing a source (S) of X-rays, a means of recording and a volume of interest hat contains the object to be X-rayed located between the source (S) and the means of recording consists of an extrapolation of each mask Mincludes determining a last segment 1beyond the limits of the means of recording; and working out a two-dimensional mask Massociated with a position Sof the source, for any angle &#x3b3; included in the angular range &#x3b8;to &#x3b8;(a position close to &#x3b8;). For every parallel segment 1 located between segment d (or d' respectively) and segment 1, the procedure further includes deriving a three-dimensional mask () of the object for each voxel at the intersection of plane Pand the volume of interest; and projecting the three-dimensional mask onto the segment 1.

Procedure for deriving a three-dimensional digital mask starting from a series of two-dimensional masks, plus a device for doing this

A method of extrapolating a bidimensional grey-level image using a radiography device.

Growing community of inventors

Chaville, France

Giovanni Palma