Method and system for crystal identification

Patent metadata

Field	Value
Publication number	US-2016321808-A1
Application number	US-201615142295-A
Country	US
Kind code	A1
Filing date	Apr 29, 2016
Priority date	Apr 29, 2015
Publication date	Nov 3, 2016
Grant date	—

Abstract

Some embodiments of the present disclosure relates to a method and system for generating crystal lookup table (CLT) based on a flood histogram. The method may include receiving a flood histogram of a subject; determining a crystal central position map based on the flood histogram, the crystal central position map including a plurality of crystal central positions, forming rows and columns of the plurality of crystal central positions to generate a labelled crystal lookup table, forming a template based on the rows and the columns in the labelled crystal central position map, and correcting the labelled crystal central position map based on the template and the flood histogram to obtain a corrected crystal central position map. A crystal lookup table may be formed.

First claim

What is claimed is:

1 . A method comprising:
receiving a flood histogram relating to a subject; determining a crystal central position map based on the flood histogram, the crystal central position map comprising a plurality of crystal central positions; forming rows and columns of the plurality of crystal central positions to generate a labelled crystal central position map; forming a template based on the rows and the columns in the labelled crystal central position map; and correcting the labelled crystal central position map based on the template to obtain a corrected crystal central position map.

2 . The method of claim 1 , the determining the crystal central position map based on the flood histogram comprising:
performing a zero-one normalization on the flood histogram to generate a normalized flood histogram I_0; performing K iterations of decay mask treatment on I_0 to generate a series of intermediate flood histogram I_i, 1≦i≦K; for each of the intermediate flood histogram I_i, performing a threshold determination on I_i to generate a binary image B_i; and generating the crystal central position map based on a group comprising the binary images B_i, 1≦i≦K.

3 . The method of claim 2 , the performing a zero-one normalization on the flood histogram comprising:
performing a linear transformation on the flood histogram to obtain a transformed flood histogram, the intensities of the transformed flood histogram falling in a range between zero and one.

4 . The method of claim 2 , the decay mask treatment on an image comprising:
multiplying the image by the normalized flood histogram I_0.

5 . The method of claim 2 , the performing a threshold determination on I_i comprising performing an Otsu thresholding algorithm on I_i.

6 . The method of claim 2 , the generating a crystal central position map comprising:
setting an initial collection of connected components C to be void; for each B_i, i=K, K−1, . . . , 2, 1,
identifying a plurality of connected components in B_i;
classifying the connected components in B_i as overlapping or non-overlapping with respect to C;
collecting the set of non-overlapping connected components in B_i into C; and
creating the crystal central position map based on C.

7 . The method of claim 6 further comprising:
for each of the connected components in C, identifying a center of mass of the connected component as a crystal central position in the flood histogram.

8 . The method of claim 1 , the determining the crystal central position map comprising calculating a local maximum gray level of the flood histogram.

9 . The method of claim 1 , the forming rows and columns comprising
performing Hough transformation on the flood histogram to generate a transformed image; selecting a configuration template; for a first crystal central position of the plurality of crystal central positions, identifying one or more candidate crystal central positions based on the selected configuration template and the transformed image; and marking the one or more candidate crystal central positions as in a same row or a same column as the first crystal central position.

10 . The method of claim 1 , the correcting the labelled crystal central position map based on the template comprising:
iteratively regularizing the template.

11 . A non-transitory computer readable medium comprising executable instructions that, when executed by at least one processor, cause the at least one processor to effectuate a method comprising:
receiving a flood histogram relating to a subject; determining a crystal central position map based on the flood histogram, the crystal central position map comprising a plurality of crystal central positions; forming rows and columns of the plurality of crystal central positions to generate a labelled crystal lookup table; forming a template based on the rows and the columns in the labelled crystal central position map; and correcting the labelled crystal central position map based on the template to obtain a corrected crystal central position map.

12 . The non-transitory computer readable medium of claim 11 , the determining the candidate crystal central position map based on the flood histogram comprising:
performing a zero-one normalization on the flood histogram to generate a normalized flood histogram I_0; performing K iterations of decay mask treatment on I_0 to generate a series of intermediate flood histogram I_i, 1≦i≦K; for each of the intermediate flood histogram I_i, performing a threshold determination on I_i to generate a binary image B_i; and
generating the crystal central position map based on a group comprising the binary images B_i, 1≦i≦K.

13 . The method of claim 12 , the performing a zero-one normalization on the flood histogram comprising:
performing a linear transformation on the flood histogram so that the range of transformed intensity is between zero and one.

14 . The method of claim 12 , the decay mask treatment on an image comprising:
multiplying the image by the normalized flood histogram.

15 . The method of claim 12 , the performing automatic threshold determination on I_i further comprising: performing an Otsu thresholding algorithm.

16 . The method of claim 12 , the generating a crystal central position map based on the binary images B_i, 1≦i≦K comprising:
setting the initial collection of connected components C to be void;
for each B_i, i=K, K−1, . . . , 2, 1,
identifying a plurality of connected components in B_i;
classifying the connected components in B_i as overlapping or non-overlapping with respect to C;
collecting the set of non-overlapping connected components in B_i into C; and

creating the crystal central position map based on C.

17 . The method of claim 6 further comprising:
for each of the connected components in C, identifying center of mass of the connected component as a crystal central position in the flood histogram associated with the connected component.

18 . The method of claim 11 , the determining the crystal central position map comprising:
calculating a local maximum gray level of the flood histogram.

19 . The method of claim 11 , the forming rows and columns comprising
performing Hough transformation on the flood histogram to generate a transformed image; selecting a configuration template; for a first crystal central position of the plurality of crystal central positions, identifying one or more candidate crystal central positions based on the selected configuration template and the transformed image; and marking the one or more candidate crystal central positions as in a same row or a same column as the first crystal central position.

20 . The method of claim 11 , the correcting the labelled crystal central position map based on the template comprising:
iteratively regularizing the template.

Assignees

• Shanghai United Imaging Healthcare Co Ltd

Inventors

• Zhou Xin
• Li Liping
• Li Qiang
• Lv Xinyu
• An Shaohui

Classifications

• G01T1/1648 Primary

  Ancillary equipment for scintillation cameras, e.g. reference markers, devices for removing motion artifacts, calibration devices (adapted for flow studies G01T1/1647) · CPC title

• G06T12/10

  Image preprocessing, e.g. calibration, positioning of sources or scatter correction · CPC title

• G06K9/4642

  Physics · mapped topic

• G06T7/0042

  Physics · mapped topic

• G06K9/20

  Physics · mapped topic