Koninklijke Philips Corporation N.v.

Andriy Andreyev

Chuanyong Bai

Zhiqiang Hu

Bin Zhang

Xiyun Song

Jinghan Ye

Yang-Ming Zhu

Shekhar Dwivedi

Chi-Hua Tung

Shushen Lin

Yu-Lung Hsieh

Andreas Goedicke

Douglas B McKnight

Andre Frank Salomon

Michael Allen Miller

Yanfei Mao

Steven Michael Cochoff

Manoj Narayanan

Leonid Romanov

Thomas Christopher Bulgrin

Dane Pittock

Nathan Serafino

Ravindra Mohan Manjeshwar

Faguo Yang

Sydney Kaplan

Thomas Christoher Bulgrin

James Gurian

Kushal Shah

Gregory Doughty

Ilya Brodskiy

A non-transitory computer-readable medium stores instructions readable and executable by at least one electronic processor () to perform an imaging method (). The method includes: reconstructing emission imaging data to generate an emission image of a lesion; converting intensity values of the emission image to at least one standardized uptake value (SUV value) for the lesion; processing input data using a regression neural network (NN) () to output an SUV correction factor for the lesion, wherein the input data includes at least two of (i) image data comprising the emission image or a feature vector representing the emission image, (ii) the at least one SUV value, (iii) a size of the lesion, and (iv) reconstruction parameters used in the reconstructing; and controlling a display device () to display at least one of (I) the SUV correction factor and (II) a corrected SUV value generated by applying the SUV correction factor to the at least one SUV value.

Artificial intelligence (AI)-based standardized uptake value (SUV) correction and variation assessment for positron emission tomography (PET)

An image reconstruction method includes: determining singles rates of radiation detectors in a frame of imaging data detected by the radiation detectors; determining an energy correction factor (N) for each radiation detector based on an energy spectrum distribution of gamma rays incident on the radiation detector during acquisition of the frame of imaging data; determining a singles live time correction factor for each radiation detector from the singles rate and the energy correction factor; determining a system coincidence live time correction factor from the system singles rate; for each line of response (LOR) connecting pairs of radiation detectors, determining a live time correction factor for the LOR from the determined singles live time correction factors of the pair of radiation detectors connected by the LOR and the determined system coincidence live time correction factor; and reconstructing the frame of imaging data using the determined LOR live time correction factors.

Dead-time correction method in quantitative positron emission tomography (PET) reconstruction for various objects and radioactivity distributions

A non-transitory computer-readable medium stores instructions readable and executable by a workstation () including at least one electronic processor () to perform an imaging method (). The method includes: receiving imaging data on a frame by frame basis for frames along an axial direction with neighboring frames overlapping along the axial direction wherein the frames include at least a volume (k) and a succeeding volume (k+1) at least partially overlapping the volume (k) along the axial direction; and generating an image of the volume (k) using an iterative image reconstruction process in which an iteration of the iterative image reconstruction process includes: computing a local penalty function for suppressing noise over the volume (k) including reducing the value of the local penalty function in an overlap region; generating an update image of the volume (k) using imaging data from the volume (k) and further using the local penalty function.

Edge preserving penalized reconstruction for step-and-shoot and motion compensated positron emission tomography (PET) studies

Imaging data () are acquired by a PET scanner () or other imaging device. Iterative image reconstruction of the imaging data is performed to generate a reconstructed image (). The iterative image reconstruction includes performing an update step () that includes an edge preserving prior () having a spatially varying edge preservation threshold () whose value at each image voxel depends on a noise metric () in a local neighborhood of the image voxel. The noise metric may be computed as an aggregation of the intensities of neighborhood image voxels of the reconstructed image in the local neighborhood of the image voxel. The edge preserving prior may be a Relative Difference Prior (RDP). For further noise suppression, during the iterative image reconstruction image values of image features of the reconstructed image that have spatial extent smaller than a threshold () may be reduced.

Information-adaptive regularization for iterative image reconstruction

A device () for measuring respiration of a patient includes a positron emission tomography (PET) or single photon emission computed tomography (SPECT) imaging device (). At least one electronic processor () is programmed to: extract a first respiration data signal () from emission imaging data of a patient acquired by the PET or SPECT imaging device; extract a second respiration data signal () from a photoplethysmograph (PPG) signal of the patient; and combine the first and second extracted respiration data signals to generate a respiration signal () indicative of respiration of the patient.

Respiratory gating using pulse oximeters for tomographic imaging

A device () for performing an amyloid assessment includes a radiation detector assembly () including at least one radiation detector (). At least one electronic processor () is programmed to: detect radiation counts over a data acquisition time interval using the radiation detector assembly; compute at least one current count metric from the detected radiation counts; store the at least one current count metric associated with a current test date in a non-transitory storage medium (); and determine an amyloid metric based on a comparison of the at least one current count metric with a count metric stored in the non-transitory storage medium associated with an earlier test date.

Method and apparatus for amyloid screening

A non-transitory computer-readable medium stores instructions readable and executable by a workstation () including at least one electronic processor () to perform an image reconstruction method (). The method includes: determining a weighting parameter () of an edge-preserving regularization or penalty of a regularized image reconstruction of an image acquisition device () for an imaging data set obtained by the image acquisition device; determining an edge sensitivity parameter (&#x3b3;) of the edge-preserving algorithm for the imaging data set obtained by the image acquisition device; and reconstructing the imaging data set obtained by the image acquisition device to generate a reconstructed image by applying the regularized image reconstruction including the edge-preserving regularization or penalty with the determined weighting and edge sensitivity parameters to the imaging data set obtained by the image acquisition device.

Image reconstruction employing tailored edge preserving regularization

An imaging method () includes: acquiring first training images of one or more imaging subjects using a first image acquisition device (); acquiring second training images of the same one or more imaging subjects as the first training images using a second image acquisition device () of the same imaging modality as the first imaging device; and training a neural network (NN) () to transform the first training images into transformed first training images having a minimized value of a difference metric comparing the transformed first training images and the second training images.

Positron emission tomography (PET) system design optimization using deep imaging

Emission imaging data are reconstructed to generate a low dose reconstructed image. Standardized uptake value (SUV) conversion () is applied to convert the low dose reconstructed image to a low dose SUV image. A neural network () is applied to the low dose SUV image to generate an estimated full dose SUV image. Prior to applying the neural network the low dose reconstructed image or the low dose SUV image is filtered using a low pass filter (). The neural network is trained on a set of training low dose SUV images and corresponding training full dose SUV images to transform the training low dose SUV images to match the corresponding training full dose SUV images, using a loss function having a mean square error loss component () and a loss component () that penalizes loss of image texture and/or a loss component () that promotes edge preservation.

Full dose PET image estimation from low-dose PET imaging using deep learning

A non-transitory computer-readable medium stores instructions readable and executable by a workstation () including at least one electronic processor () to perform an image reconstruction method () to reconstruct list mode data acquired over a frame acquisition time using a plurality of radiation detectors () in which the events of the list mode data is timestamped. The method includes: for the sub-frame bins of a plurality of sub-frame bins into which the frame acquisition time is divided, determining a sub-frame singles rates map for the plurality of radiation detectors from the list mode data whose time stamps reside in the sub-frame bin; determining a singles rate for the singles events of the list mode data using the sub-frame singles rates maps wherein the singles rates for the singles events are determined at a temporal resolution that is finer than the frame acquisition time; determining correction factors for the list mode data using the determined singles rates for the singles events of the list mode data; and reconstructing the list mode data of the frame acquisition time using the determined correction factors to generate a reconstructed image for the frame acquisition time.

Correction method for quantification accuracy improvement in list mode reconstruction

A non-transitory computer-readable medium stores instructions readable and executable by a workstation () operatively connected to a display device () and including at least one electronic processor () to perform an image acquisition and reconstruction method (). The method includes: retrieving a non-voxel-based reconstructed image comprising non-voxel image elements from a picture and archiving communication system (PACS) database () to the workstation; at the workstation, generating at least one voxel-based resampled image from the non-voxel-based reconstructed image; and displaying the at least one voxel-based reconstructed image on the display device.

Real time reconstruction-native image element resampling for high definition image generation and processing

A non-transitory storage medium stores instructions readable and executable by an electronic processor () to perform a method () for estimating singles rates for detectors () of a detector array () of a positron emission tomography (PET) imaging device (). The method includes: obtaining measured delayed coincidence rates for lines of response (LORs) defined by pairs of the detectors of the detector array; and generating estimated singles rates for the detectors of the detector array by an iterative optimization in which each iteration performs an update of estimates of the singles rates for the detectors by operations including: forward projecting the estimates of the singles rates for the detectors to generate estimates of delayed coincidence rates for the LORs; generating quantitative comparisons between the measured delayed coincidence rates for the LORs and the generated estimates of the delayed coincidence rates for the LORs; backprojecting the quantitative comparisons to generate corrections for the singles rates for the detectors; and updating the estimates of the singles rates for the detectors using the generated corrections. The estimated singles rates for the detectors of the detectors array comprise the estimates of the singles rates for the detectors output by the iterative optimization when the iterative optimization satisfies a stopping criterion. This process jointly optimizes the estimated singles rates and randoms estimation.

Relaxed iterative maximum-likelihood expectation maximization for positron emission tomography random coincidence estimation

A non-transitory computer readable medium storing instructions readable and executable by an imaging workstation () including at least one electronic processor () to perform a dataset generation method () operating on emission imaging data acquired of a patient for one or more axial frames at a corresponding one or more bed positions, the method comprising: (a) identifying a frame of interest from the one or more axial frames; (b) generating simulated lesion data by simulating emission imaging data for the frame of interest of at least one simulated lesion placed in the frame of interest; (c) generating simulated frame emission imaging data by simulating emission imaging data for the frame of interest of the patient; (d) determining a normalization factor comprising a ratio of the value of a quantitative metric for the simulated patient data and the value of the quantitative metric for the emission imaging data acquired of the same patient for the frame of interest; and (e) generating a hybrid data set comprising the emission imaging data acquired of the patient for the one or more axial frames at the corresponding one or more bed positions with the frame of interest replaced by a combination of the simulated lesion data scaled by the normalization factor and the emission imaging data acquired of the patient for the frame of interest.

Generation of accurate hybrid datasets for quantitative molecular imaging

A non-transitory computer-readable medium stores instructions readable and executable by at least one electronic processor () to perform an image reconstruction method (). The method includes: performing iterative image reconstruction of imaging data acquired using an image acquisition device (); selecting an update image from a plurality of update images produced by the iterative image reconstruction; processing the selected update image to generate a hot spot artifact map; and suppressing hot spots identified by the generated hot spot artifact map in a reconstructed image output by the iterative image reconstruction.

Learning-based voxel evolution for regularized reconstruction

A non-transitory computer-readable medium stores instructions executable by a processor to perform an acquisition and reconstruction method for a first image acquisition device. The method includes determining a scheduled acquisition time based on an attenuation map derived from imaging data from a second image acquisition device and a sensitivity matrix of the first image acquisition device; acquiring emission imaging data using the first image acquisition device, where the acquiring is scheduled to be performed over the scheduled acquisition time; during an initial portion of the acquiring, measuring a count or count rate of the acquired emission imaging data; adjusting the scheduled acquisition time based on the measured count or count rate to generate an adjusted acquisition time while continuing the acquiring; stopping the acquiring at the adjusted acquisition time; and reconstructing the emission imaging data acquired over the adjusted acquisition time to generate one or more reconstructed images.

Growing community of inventors

Willoughby Hills, OH, United States of America

Andriy Andreyev