General Electric Company

Ge Medical Systems Global Technology Company, LLC

University of Notre Dame Du Lac

Purdue Research Foundation

Ge Precision Healthcare LLC

Other

Wisconsin Alumni Research Foundation

University of Michigan

University of Iowa Research Foundation

Rensselaer Polytechnic Institute

General Electric Corporation Gmbh

Jiang Hsieh

Jean-Baptiste Thibault

Robert F Senzig

Ken David Sauer

Charles Addison Bouman

Gopal Biligeri Avinash

Thomas L Toth

Jianying Li

Xiangyang Tang

Saad Ahmed Sirohey

Xiaoye Wu

Darin Robert Okerlund

Bruno Kristiaan Bernard De Man

Baojun Li

Stanley Haim Fox

Sandeep Dutta

Brian Edward Nett

Jed Douglas Pack

Naveen Stephan Chandra

Peter Michael Edic

Jiahua Fan

Kishore Chandra Acharya

Paavana Sainath

Melissa Liliane Vass

John Howard Londt

Stephen Wayne Metz

Mark Edward Woodford

Fang Frank Dong

Ge Wang

Samit Kumar Basu

John Michael Sabol

Guang-Hong Chen

Abdelaziz Ikhlef

Dan Xu

Zhou Yu

Bruno De Man

Bernice Hoppel

Debashish Pal

Hui Hu

Zhye Yin

Xin Wang

Bipul Das

Cherik T Bulkes

Paul Edgar Licato

Yasuhiro Imai

Kai Zeng

Tin-Su Pan

Roy A Nilsen

Bernhard Erich Hermann Claus

Michael F Gard

Jerome F Knoplioch

Guy M Besson

Anne Lindsay Hall

Clarence Lavere Gordon, Iii

Masatake Nukui

Bruce Matthew Dunham

Jonathan Richard Schmidt

Yannan Jin

Ricardo Scott Avila

Lin Fu

Giridhar Jothiprasad

Rahul Bhotika

Ping Liu

Phil E Pearson, Jr

Steven James Gray

Hui David He

Jeffrey Allen Fessler

Steven John Woloschek

John Irvin Jackson

Jiaqin Dong

Maria Iatrou

Shuo Li

David Michael Hoffman

Yaron Hefetz

John Eric Tkaczyk

Kevin Franklin King

Jean-Paul Bouhnik

Akira Hagiwara

Daniel David Harrison

Michael J Washburn

Charles William Stearns

John W Newman

Carl R Crawford

James Vradenburg Miller

David Allen Langan

James Walter LeBlanc

Nicholas Ryan Konkle

Jeffrey Alan Kautzer

Alexander Ganin

Alexander Katsevich

Bing A Shen

Xueli Wang

Hao Lai

Erik Normann Steen

Jason Stuart Katcha

Shankara Bonthu Reddy

Mark Vincent Profio

Prem Venugopal

Armin H Pfoh

Bruno Kristiaan Bernard DeMan

Techniques are described for enhancing the quality of three-dimensional (3D) anatomy scan images using deep learning. According to an embodiment, a system is provided that comprises a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory. The computer executable components comprise a reception component that receives a scan image generated from 3D scan data relative to a first axis of a 3D volume, and an enhancement component that applies an enhancement model to the scan image to generate an enhanced scan image having a higher resolution relative to the scan image. The enhancement model comprises a deep learning neural network model trained on training image pairs respectively comprising a low-resolution scan image and a corresponding high-resolution scan image respectively generated relative to a second axis of the 3D volume.

Deep learning-based image quality enhancement of three-dimensional anatomy scan images

Systems and methods are provided for increasing a quality of computed tomography (CT) images. In one embodiment, a computed tomography (CT) detector system comprises a layer of energy integrating detectors (EID) arranged below a layer of photon counting (PC) sensors with respect to an incoming x-ray, where a number of the PC sensors exceeds a number of the EID detectors; and an image processing unit configured to correct PC data using EID data, and denoise and increase a resolution of an image reconstructed from EID data and PC data using a deep learning convolutional neural network (CNN) trained on pairs of images, each pair of images including a target image reconstructed from a first signal from the layer of PC sensors, and an input image reconstructed from a second signal from the layer of EID detectors, the EID data and PC data acquired concurrently from a same patient ray path.

Systems and methods for computed tomography

Techniques are described for automatically detecting scan characteristics of a medical image series. According to an embodiment, a system is provided that comprises a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory. The computer executable components comprise an image generation component that generates a representative image of a medical image series comprising a plurality of scan images, and a series characterization component that processes the representative image using one or more characteristic detection algorithms to determine one or more characteristics of the medical image series. The system can further tailor the visualization layout for viewing the medical image series based on the one or more characteristics and/or automatically perform various workflow tasks based on the one or more characteristics.

Automatically detecting characteristics of a medical image series

Methods and systems are provided for identifying motion in medical images. In one example, a method includes obtaining projection data of an imaging subject, reconstructing a first image of a location of the imaging subject from the projection data using a first reconstruction technique and reconstructing a second image corresponding to the same location of the imaging subject from the of projection data using a second reconstruction technique, different than the first reconstruction technique in terms of temporal sensitivity, calculating an inconsistency metric quantifying temporal inconsistencies between the first image and the second image, and taking an action based on the inconsistency metric.

Systems and methods for motion detection in medical images

Various methods and systems are provided for computed tomography imaging. In one embodiment, a method includes acquiring, with an x-ray detector and an x-ray source coupled to a gantry, a three-dimensional image volume of a subject while the subject moves through a bore of the gantry and the gantry rotates the x-ray detector and the x-ray source around the subject, inputting the three-dimensional image volume to a trained deep neural network to generate a corrected three-dimensional image volume with a reduction in aliasing artifacts present in the three-dimensional image volume, and outputting the corrected three-dimensional image volume. In this way, aliasing artifacts caused by sub-sampling may be removed from computed tomography images while preserving details, texture, and sharpness in the computed tomography images.

Method and systems for aliasing artifact reduction in computed tomography imaging

Systems/techniques that facilitate machine learning generation of low-noise and high structural conspicuity images are provided. In various embodiments, a system can access an image and can apply at least one of image denoising or image resolution enhancement to the image, thereby yielding a first intermediary image. In various instances, the system can generate, via execution of a plurality of machine learning models, a plurality of second intermediary images based on the first intermediary image, wherein a given machine learning model in the plurality of machine learning models receives as input the first intermediary image, wherein the given machine learning model produces as output a given second intermediary image in the plurality of second intermediary images, and wherein the given second intermediary image represents a kernel-transformed version of the first intermediary image. In various cases, the system can generate a blended image based on the plurality of second intermediary images.

Machine learning generation of low-noise and high structural conspicuity images

Methods and systems are provided for increasing a quality of computed tomography (CT) images reconstructed from high helical pitch scans. In one embodiment, the current disclosure provides for a method comprising generating a first computed tomography (CT) image from projection data acquired at a high helical pitch; using a trained multidimensional statistical regression model to generate a second CT image from the first CT image, the multidimensional statistical regression model trained with a plurality of target CT images reconstructed from projection data acquired at a lower helical pitch; and performing an iterative correction of the second CT image to generate a final CT image.

Systems and methods for computed tomography image reconstruction

A computer-implemented method for correcting artifacts in computed tomography data is provided. The method includes inputting a sinogram into a trained sinogram correction network, wherein the sinogram is missing a pixel value for at least one pixel. The method also includes processing the sinogram via one or more layers of the trained sinogram correction network, wherein processing the sinogram includes deriving complementary information from the sinogram and estimating the pixel value for the at least one pixel based on the complementary information. The method further includes outputting from the trained sinogram correction network a corrected sinogram having the estimated pixel value.

System and method for utilizing a deep learning network to correct for a bad pixel in a computed tomography detector

Various methods and systems are provided for x-ray tube conditioning for a computed tomography imaging method. In one embodiment, x-ray may be generated in an x-ray tube of a radiation source prior to a diagnostic scan to warmup the x-ray tube to a desired temperature for the diagnostic scan. The power delivered to the x-ray tube during warmup may be adjusted in a closed loop system based on an initial temperature of the x-ray tube and the desired temperature for the diagnostic scan. During tube warmup, by placing a blocking plate coupled to a collimator blade in a path of the x-ray beam, the x-ray beam may be blocked from exiting a collimator.

Methods and systems for x-ray tube conditioning

A system and method for estimating vascular flow using CT imaging include a computer readable storage medium having stored thereon a computer program comprising instructions, which, when executed by a computer, cause the computer to acquire a first set of data comprising anatomical information of an imaging subject, the anatomical information comprises information of at least one vessel. The instructions further cause the computer to process the anatomical information to generate an image volume comprising the at least one vessel, generate hemodynamic information based on the image volume, and acquire a second set of data of the imaging subject. The computer is also caused to generate an image comprising the hemodynamic information in combination with a visualization based on the second set of data.

System and method for estimating vascular flow using CT imaging

The present disclosure relates to image reconstruction with favorable properties in terms of noise reduction, spatial resolution, detail preservation and computational complexity. The disclosed techniques may include some or all of: a first-pass reconstruction, a simplified datafit term, and/or a deep learning denoiser. In various implementations, the disclosed technique is portable to different CT platforms, such as by incorporating a first-pass reconstruction step.

Iterative image reconstruction framework

A system for reducing artifact bloom in a reconstructed image of an object is provided. The system includes an imaging device, and a controller. The imaging device is operative to obtain one or more slices of the object. The controller is in electronic communication with the imaging device and operative to: generate the reconstructed image based at least in part on the one or more slices; and de-bloom one or more regions within the reconstructed image based at least in part on a contrast medium enhancement across at least part of a volume of the object.

System and method for reducing artifact bloom in a reconstructed object

Methods and systems for X-ray tube conditioning

The present approach relates to the training of a machine learning algorithm for image generation and use of such a trained algorithm for image generation. Training the machine learning algorithm may involve using multiple images produced from a single set of tomographic projection or image data (such as a simple reconstruction and a computationally intensive reconstruction), where one image is the target image that exhibits the desired characteristics for the final result. The trained machine learning algorithm may be used to generate a final image corresponding to a computationally intensive algorithm from an input image generated using a less computationally intensive algorithm.

Image generation using machine learning

A method relates to the use of deep learning techniques, which may be implemented using trained neural networks (), to estimate various types of missing projection or other unreconstructed data. Similarly, the method may also be employed to replace or correct corrupted or erroneous projection data as opposed to estimating missing projection data.

Deep learning based estimation of data for use in tomographic reconstruction

Growing community of inventors

Brookfield, WI, United States of America

Jiang Hsieh