Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

ENHANCED NON-RIGID REGISTRATION UTILIZING A QUIESCENT-PERIOD REFERENCE GATE FOR POSITRON EMISSION TOMOGRAPHY IMAGE RECONTRUCTION

IP.com Disclosure Number: IPCOM000239687D
Publication Date: 2014-Nov-25
Document File: 5 page(s) / 176K

Publishing Venue

The IP.com Prior Art Database

Abstract

The present invention proposes a technique to utilize acquired data which contributes to an improved final positron emission tomography (PET) image volume. The technique includes regularized image reconstruction to allow smooth image data of gates nearby to a quiescent gate. The technique includes deriving non-rigid registration (NRR) motion vector on a coarse grid from smooth image data and applying to standard image grid gate data. This technique is optionally applied when the registered data increases signal to noise ratio (SNR) of a user-defined feature-of-interest and adds to the final image. The technique includes two embodiments to improve final image volume.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 44% of the total text.

ENHANCED NON-RIGID REGISTRATION UTILIZING A QUIESCENT-PERIOD REFERENCE GATE FOR POSITRON EMISSION TOMOGRAPHY IMAGE RECONTRUCTION

FIELD OF INVENTION

The invention generally relates to a positron emission tomography (PET) image reconstruction and more particularly to an improved final image volume of the PET image.

BACKGROUND OF THE INVENTION

Generally, respiratory motion degrades positron emission tomography (PET) image quality and quantitative accuracy. In order to mitigate impact of respiratory motion, images utilize only coincidence data acquired during quiescent portion of the respiratory cycles.  In computer tomography (PET/CT) practice, utilization of such quiescent period coincidence data is up to approximately 50% of the acquired data. Increase in total amount of coincidence data that contributes to the final PET image would enhance image quality due to noise reduction. Such data is required to be adequately corrected to the most motion-free portion of the data or the quiescent portion.

A conventional technique includes upper and lower amplitude thresholds per respiratory cycle to define a quiescent portion of data. Another conventional technique includes magnetic resonance (MR) navigation information to find motion and enable a motion correction. Yet another conventional technique includes utilizing computer tomography (CT) to adjust or define PET motion correction.

However, the conventional techniques do not acquire data which contributes to an improved final image volume.

 It would be desirable to have an efficient technique to acquire data which contributes to an improved final image volume.

BRIEF DESCRIPTION OF THE INVENTION

The present invention proposes a technique to utilize acquired data which contributes to an improved final positron emission tomography (PET) image volume. The technique includes regularized PET image reconstruction to allow increased signal-to-noise PET image data of gates nearby to a quiescent gate. Regularization within the PET reconstruction enables increased feature contrast at equivalent image noise or similar feature contrast at reduced background noise – either can give the impression of a smoother image. The technique includes deriving non-rigid registration (NRR) motion vector on a coarse grid from the regularized image data and applying to standard image grid gate data.

DETAILED DESCRIPTION OF THE INVENTION

The present invention proposes a technique to utilize acquired data which contributes to an improved final positron emission tomography (PET) image volume. The technique includes regularized image reconstruction to allow smooth image data of gates nearby to a quiescent gate. The technique includes deriving non-rigid registration (NRR) motion vector on a coarse grid from smooth image data and applying to standard image grid gate data. This technique is optionally applied when the registered data increases signal to noise ratio (SNR) of a user-defined feature-of-interest and adds to the final ima...