Koninklijke Philips Corporation N.V.

Johns Hopkins University

Patrick Olivier

Bin Zhang

Amy Perkins

Parmeshwar Khurd

Daniel Gagnon

Joerg Bredno

Sven Prevrhal

Chi-Hua Tung

Andreas Goedicke

Andre Frank Salomon

Jeffrey Allan Kolthammer

Piotr J Maniawski

Benjamin Tsui

Parmeshwar Kishore Khurd

A respiratory motion estimation method () includes reconstructing emission imaging data () to generate a reconstructed image (). The emission imaging data comprises lines of response (LORs) acquired by a positron emission tomography (PET) imaging device or projections acquired by a gamma camera. One or several assessment volumes () are defined within the reconstructed images. The emission imaging data are binned into time interval bins based on time stamps of the LORs or projections. A displacement versus time curve () is generated by computing, for each time interval bin, a statistical displacement metric of the LORs or projections that both are binned in the time interval bin and intersect the motion assessment volume. The motion assessment volume may be selected to overlap a motion assessment image feature () identified in the reconstructed image.

Methods for data driven respiratory motion estimation

A respiratory motion signal generation method operates on emission data () of an imaging subject in an imaging field of view (FOV) acquired by a positron emission tomography (PET) or single photon emission computed tomography (SPECT) imaging device (). An array of regions () is defined in the imaging FOV without reference to anatomy of the imaging subject. For each region of the array of regions defined in the imaging FOV, an activity position versus time curve () is computed from the emission data acquired by the PET or SPECT imaging device. Frequency-selective filtering of the activity position versus time curves is performed to generate filtered activity position versus time curves. At least one motion signal () is generated by combining the filtered activity position versus time curves of at least a selected sub-set of the regions.

Noise-robust real-time extraction of the respiratory motion signal from PET list-data

A nuclear imaging apparatus () acquires nuclear imaging data comprising events wherein each event records at least spatial localization information and a timestamp for a nuclear decay event. An event-preserving image reconstruction module () reconstructs the nuclear imaging data using an event-preserving reconstruction algorithm to generate an image represented as an event-preserving reconstructed image dataset (I) comprising for each event the timestamp and at least one spatial voxel assignment. One or more structures are identified in the image and independent motion compensation is performed for each structure. In one approach, an events group is identified corresponding to the structure comprising events assigned to the structure by the event-preserving reconstructed image dataset; a time binning of the events of each events group is optimized based on a motion profile for the structure; time bin images are generated; and the structure is spatially registered in the time bin images.

List mode dynamic image reconstruction

An image processing apparatus includes a geometric correction memory which stores a pre-calculated geometric matrix specific to an imaging apparatus. An attenuation map memory which stores an attenuation map of a subject to be imaged in the imaging apparatus. A buffer stores a plurality of lines generated by the imaging apparatus to be reconstructed. A processor reconstructs the lines into a attenuation corrected image representation of the subject using the lines from the buffer, the attenuation map, and the geometric matrix.

Fast computation of the sensitivity matrix in iterative algorithms

A PET scanner () generates a plurality of time stamped lines of response (LORs). A motion detector () detects a motion state, such as motion phase or motion amplitude, of the subject during acquisition of each of the LORs. A sorting module () sorts the LORs by motion state and a reconstruction processor () reconstructs the LORs into high spatial, low temporal resolution images in the corresponding motion states. A motion estimator module () determines a motion transform which transforms the LORs into a common motion state. A reconstruction module () reconstructs the motion corrected LORs into a static image or dynamic images, a series of high temporal resolution, high spatial resolution images.

Method and apparatus to detect and correct motion in list-mode PET data with a gated signal

In accordance with one aspect of the invention a method and apparatus for utilizing time of flight information to detect motion during a medical imaging acquisition, such as a PET imaging acquisition, is provided. In accordance with another aspect of the invention, a method and apparatus for detecting and correcting for respiratory motion and cardiac motion in medical images, such as PET images, is provided.

Method and apparatus for using time of flight information to detect and correct for motion in imaging scans

A continuous time-of-flight scatter simulation method is provided, with a related method of correcting PET imaging data to compensate for photon scatter. Scatter contributions from each imaging pixel in a field of view may be calculated without binning data.

Continuous time-of-flight scatter simulation method

A diagnostic imaging device includes detector elements () for detecting &#x3b3;-rays indicative of nuclear decay events. The detected &#x3b3;-rays are used to produce lines of response (LORs) (), which are time stamped () and stored in list mode. The LORs are reconstructed () into an image. An image analysis processor () analyzes the image for motion artifacts and iteratively adjusts an event transform processor () to transform selected LORs to minimize the motion artifacts. If the transformed LOR () does not correspond with a pair of detector elements (), closest detector elements () are determined. Candidate LORs () are created between the closest and neighboring detector elements. An event location () on an LOR () is determined from the time-of-flight (TOF) information and then transformed () to generate a transformed event location (). The candidate LOR () which most nearly intersects the transformed event location () and the appropriately updated TOF information is selected for use in image reconstruction.

Growing community of inventors

Solon, OH, United States of America

Patrick Olivier