Browse Prior Art Database

ADAPTIVE ACQUISITION AND RECONSTRUCTION PARAMETER SETTING ENGINE

IP.com Disclosure Number: IPCOM000247674D
Publication Date: 2016-Sep-27
Document File: 5 page(s) / 202K

Publishing Venue

The IP.com Prior Art Database

Abstract

A technique to data mine user preferences in order to modify acquisition and image reconstruction parameters on an individual or institution basis for a Computed Tomography (CT) scanner is proposed. The technique provides an automated means of receiving multiple forms of user input including a structured rating on Picture Archiving and Communication System (PACS) in order to provide optimal acquisition and reconstruction parameters on an individual and institution level. The technique uses a machine learning algorithm to select the image reconstruction parameters, where the user manually enters their satisfaction through rating criteria. A system records and uses information such as the duration of reading, and accuracy of the reading. The reconstruction parameters are adapted based on a training procedure. The training procedure starts with an initial set of images reviewed by each user or set of users in order to establish the initial parameters.

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 40% of the total text.

ADAPTIVE ACQUISITION AND RECONSTRUCTION PARAMETER SETTING ENGINE

BACKGROUND

The present invention relates generally to a computed tomography (CT) scanner and more particularly to a technique for adaptive acquisition and reconstruction parameter setting on an individual user or institution basis for the CT scanner.

A standard procedure for setting image reconstruction and image acquisition parameters on for a Computed Tomography (CT) scanner are typically set by an institution by either a technologist, a lead technologist, a radiologist, a physicist or a combination of one or more of these individuals. Output of the CT scanner depends upon a user to modify parameters. In reading CT images, there are preferences for image noise, resolution and texture. The preferences are not directly taken into account as it is difficult for an individual user or institution to optimize image reconstruction parameters.

Therefore, it would be desirable to have a technique to optimize image reconstruction parameters for the individual user or institution.

BRIEF DESCRIPTION OF DRAWINGS

Figure 1 depicts a sample workflow for adaptive acquisition and reconstruction parameter setting engine wherein different user input is iteratively entered and the system continues to adapt the acquisition and reconstruction settings based on the user preferences.

Figure 2 depicts a sample workflow for the adaptive acquisition and reconstruction parameter setting engine wherein information from a scheduling client is available.

Figure 3 depicts a sample workflow for the adaptive acquisition and reconstruction parameter setting engine, wherein the user inputs, for example, that of a physicist, a technologist, and a radiologists are all taken into account and the system requests that the CT console generate multiple basis images which are further combined on an advanced workstation (AW) or PACS workstation.

DETAILED DESCRIPTION

A technique to data mine user preferences in order to modify acquisition and image reconstruction parameters on an individual or institution basis for a Computed Tomography (CT) scanner is proposed. The technique provides an automated means of receiving multiple forms of user input including a structured rating on Picture Archiving and Communication System (PACS) in order to provide optimal acquisition and reconstruction parameters on an individual and institution level. The image reconstruction parameters include reconstruction kernel, iterative reconstruction parameters, low signal processing and other such parameters.

The image reconstruction parameters on a CT console are set based on a reader scheduling system and selected based on an optimization function for an institution.

The technique uses a machine learning algorithm to select the image reconstruction parameters, where the user manually enters their satisfaction through rating criteria. A system records and uses information such as the duration of reading, and accuracy of the reading.

While the tech...