DENSITY GUIDED ATTENUATION MAP GENERATION IN PET/MR SYSTEMS

• US 20160320466A1
• Filed: 12/10/2014
• Published: 11/03/2016
• Est. Priority Date: 12/20/2013
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips N.V.
• Inventors: DWIVEDI, Shekhar, BERKER, Yannick, SCHULZ, Volkmar, SHAO, Lingxiong
• Granted Patent: US 10,386,439 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: A61B 5/0035 adapted for acquisition of images from more than one imaging mode, e.g. combining MRI and optical tomography
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: G01R 33/481 MR combined with positron emission tomography [PET] or single photon emission computed tomography [SPECT]
• CPC Class Codes: G01R 33/4816 NMR imaging of samples with ultrashort relaxation times such as solid samples, e.g. MRI using ultrashort TE [UTE], single point imaging, constant time imaging
• CPC Class Codes: G01R 33/4835 of multiple slices
• CPC Class Codes: G01R 33/50 based on the determination of relaxation times , e.g. T1 measurement by IR sequences; T2 measurement by multiple-echo sequences
• CPC Class Codes: G01R 33/5608 Data processing and visualization specially adapted for MR, e.g. for feature analysis and pattern recognition on the basis of measured MR data, segmentation of measured MR data, edge contour detection on the basis of measured MR data, for enhancing measured MR data in terms of signal-to-noise ratio by means of noise filtering or apodization, for enhancing measured MR data in terms of resolution by means for deblurring, windowing, zero filling, or generation of gray-scaled images, colour-coded images or images displaying vectors instead of pixels image data processing or generation, in general G06T
• CPC Class Codes: G01R 33/56509 due to motion, displacement or flow, e.g. gradient moment nulling G01R33/567 takes precedence
• CPC Class Codes: G01T 1/1603 with a combination of at least two different types of detector see provisionally also G01T1/16
• CPC Class Codes: G01T 1/1647 Processing of scintigraphic data not related to a particular imaging system G01T1/2992
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/20132 Image cropping
• CPC Class Codes: G06T 2207/30061 Lung
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/136 involving thresholding

Abstract

A lung segmentation processor (40) is configured to classify magnetic resonance (MR) images based on noise characteristics. The MR segmenatation processor generates a lung region of interest (ROI) and detailed structure segmentation of the lung from the ROI. The MR segmentation processor performs an iterative normalization and region definition approach that captures the entire lung and the soft tissues within the lung accurately. Accuracy of the segmentation relies on artifact classification coming inherently from MR images. The MR segmentation processor (40) correlates segmented lung internal tissue pixels with the lung density to determine the attenuation coefficients based on the correlation. Lung densities are computed using MR data obtained from imaging sequences that minimize echo and acquisition times. The densities differentiate healthy tissues and lesions, which an attenuation map processor (36) uses to create localized attenuation maps for the lung.

Claims

1. A lung module system, comprising:

a volume normalization module configured to normalize MR image data that expands intensity values of near zero intensity pixels and compresses intensity values of higher intensity pixels;

a thresholding module configured to generate and apply a threshold for the MR data to differentiate and separate lung pixels from non-lung pixels in a binary volume in three-dimensions;

a lung region of interest (ROI) module generating a lung ROI from the thresholded MR data;

a cropping module configured to crop the initial volume of MR data according to the lung ROI; and

a segmentation module segmenting the MR data to differentiate a lung comprising at least two of lung pixels, lesion pixels, or air pixels.

2. The system according to claim 1, further including:

a density module assessing lung tissue densities for the pixels and segregate lesion pixels from lung pixels; and

an attenuation module generating an attenuation map having accurate attenuation coefficients based on the segregated pixels.

3. The system according to claim 2, wherein:

the density module is configured to minimize the echo time and acquisition time to compute lung tissue densities.

4. The system according to claim 2, wherein:

the density module is configured to compute lung tissue densities using ultra-short echo time magnetic resonance images.

5. The system according to claim 2, wherein:

the density module is configured to compute lung tissue densities using multiple echoes for normalization.

6. The system according to claim 1, further including:

a classification module configured to classify artifacts in the MR data;

a conversion module configured to convert the lung ROI from 3D to 2D;

a ROI shape module configured to generate a lung ROI shape;

a balancing module configured to balance and tuning the lung ROI; and

a merging module configured to merge at least two lung ROIs.

7. The system according to claim 1, further including:

a conversion module configured to convert the lung ROI from three-dimensions to two-dimensional slices;

a lung box module configured to compute a lung box;

a slice module configured to compute box limits for an imaging slice; and

a tuning module configured to tune an abdomen or a trachea with respect to a lung.

9. The system according to claim 5, wherein the one or more processors is further configured or programmed to:

compute a noise characterization of the normalized volume to create a normalization ratio; and

use the normalization ratio in generating the lung ROI.

10. The system according to claim 5, wherein the one or more processors is further configured or programmed to:

classify the MR data using MR artifact classification; and

use the MR artifact classification in generating the lung ROI.

11. The system according to claim 5, wherein the one or more processors is further configured or programmed to perform a 2D connected component and relabeling analysis to convert a binary volume from three dimensions to two dimensional slices.

12. The system according to claim 5, wherein the one or more processors generates the threshold using a histogram of normalized pixel intensities.

13. The system according to claim 9, wherein the one or more processors is further configured or programmed to generate the lung ROI by:

identifying a start and an end slice of the lung;

identifying a slice between the start and end slice that includes the highest number of lung pixels; and

balancing slices before the start slice and/or after the end slice to remove trachea pixels and/or abdomen pixels.

14. The system according to claim 5, wherein the one or more processors is further configured or programmed to:

assess lung tissue densities for the pixels to segregate lesion pixels from lung pixels; and

generate an attenuation map having accurate attenuation coefficients based on the segregated pixels.

16. The method according to claim 13, further including:

computing a noise characterization of the normalized volume to create a normalization ratio; and

using the normalization ratio as a factor in generating the lung ROI.

17. The method according to claim 13, further including:

classifying the MR data using MR artifact classification; and

using the MR artifact classification as a factor in generating the lung ROI.

18. The method according to claim 13, further including:

computing lung tissue densities to segregate the lesion pixels from the lung pixels; and

generating an attenuation map having accurate attenuation coefficients based on the segregated pixels.

8. A lung segmentation system, comprising:

one or more processors configured and programmed to;

receive a volume of magnetic resonance (MR) image data;

apply a volume normalization function to the MR data to expand intensity values of near zero intensity pixels and compresses intensity values of higher intensity pixels;

generate and apply a threshold for the MR data to differentiate and separate lung pixels from non-lung pixels in a binary volume in three-dimensions;

generate a lung region of interest (ROI) from the thresholded MR data;

crop the initial volume of MR data according to the lung ROI; and

segment the MR data to differentiate a lung comprising at least two of lung pixels, lesion pixels, or air pixels.

15. A lung segmentation method, comprising:

applying a volume normalization function to a MR image to create a normalized volume image;

generating and applying a threshold for the MR image to differentiate and separate lung pixels from non-lung pixels in a binary volume in three-dimensions;

determining a lung region of interest (ROI) from the thresholded MR image;

cropping the MR image according to the lung ROI; and

segmenting the MR image to differentiate a lung comprising at least two of lung pixels, lesion pixels, or air pixels.

19. A method for computing lung densities from MR images, comprising:

scanning a lung region of interest using two or more echoes having different echo times to acquire MR data;

processing the MR data;

calculating lung densities from the MR data; and

generating an attenuation map from the lung densities.

20. A system to generate attenuation maps using MR images comprising:

a segmentation processor having a processor configured to;

segment MR images using ROI based localized segmentation accommodating for patient characteristics, image noise, artifact characteristics, and anatomical characters of a ROI;

localize at least one organ lesion using density based mapping of intensity values;

generate localized attenuation maps based on the density mapping generate an ROI;

remove artifacts within the ROI using object tracking across localized regions wherein the object tracking is adapted according to specific MR image characteristics and patient characteristics;

test the ROI for correctness; and

iteratively repeat ROI generation to optimize the generated ROI.