Methods and systems for identifying spine or bone regions in computed tomography image sequence

Patent metadata

Field	Value
Publication number	US-10181191-B2
Application number	US-201515323035-A
Country	US
Kind code	B2
Filing date	Oct 31, 2015
Priority date	Dec 2, 2014
Publication date	Jan 15, 2019
Grant date	Jan 15, 2019

Abstract

A method and system for image processing is provided. The technique includes acquiring data related to the processing, performing a pre-processing of the data, performing a segmentation of a subject, performing a post-processing of the result of the segmentation and managing storage of the data. The post-process includes a calibrating and a rendering of the data.

First claim

What is claimed is:

1. A method for Computed Tomography (CT) image processing, comprising:
a) retrieving a CT image sequence and segmenting a plurality of bone regions of the CT image sequence, the CT image sequence including one or more image slices;
b) in each image slice of the CT image sequence, determining a bone region with the maximum area among the plurality of bone regions as a spine region;
c) in each image slice of the CT image sequence, determining at least one anchor point in the top row of the spine region, and validating the at least one anchor point based on information associated with a position of the at least one anchor point in the image slice; and
d) calibrating the at least one anchor point based on the validation result,
wherein the validating the at least one anchor point comprises:
determining a y coordinate of the at least one anchor point is greater than ¼ of the height of the image slice of the CT image sequence, and a shift of an x coordinate from the central position is less than ¼ of a width of the image slice of the CT image sequence; and
determining that the at least one anchor point is valid.

2. The method of claim 1 , the a) further comprising:
retrieving the CT image sequence;
selecting a preliminary spine region in each image slice of the CT image sequence; and
pre-processing the preliminary spine region to segment the plurality of bone regions.

3. The method of claim 2 , the selecting a preliminary spine region comprising:
selecting and segmenting the CT image sequence based on information associated with a CT scan that generates the CT image sequence.

4. The method of claim 2 , the pre-processing the preliminary spine region comprising:
preforming a threshold segmentation on the preliminary spine region to identify the plurality of bone regions; and
eliminating deficiency and disconnections of the plurality of bone regions based on a morphology algorithm.

5. The method of claim 1 , the validation result comprising that the at least one anchor point has an abnormal coordinate or lacks a coordinate, and the calibrating the at least one anchor point based on the validation result comprising:
calibrating an image slice of the CT image sequence whose anchor point has an abnormal coordinate; or
calibrating an image slice of the CT image sequence whose anchor point lacks a coordinate.

6. The method of claim 5 , the calibrating an image slice of the CT image sequence whose anchor point has an abnormal coordinate comprising:
positioning the at least one anchor point of at least one image slice of the CT image sequence sequentially;
defining a sliding window;
positioning the at least one anchor point to be calibrated in the sliding window sequentially;
calculating a position mean value of the at least one anchor point in the sliding window; and
defining a first Euclidean distance between the current anchor point and the position mean value as DM and a second Euclidean distance between the current anchor point and the previous anchor point as DP, wherein DM or DP is larger than a threshold TH,
the coordinate of the current anchor point is substituted by a coordinate corresponding to the smaller of DM and DP.

7. The method of claim 5 , the calibrating an image slice of the CT image sequence whose anchor point lacks a coordinate comprising:
fixing the anchor point that lacks a coordinate based on an anchor point in an adjacent image slice of the CT image sequence.

8. A system for Computed Tomography (CT) image processing, comprising:
at least one processor, and at least one non-transitory computer-readable storage medium storing a set of instructions, the set of instructions, when executed by the at least one processor, cause the system to:
a) retrieve a CT image sequence and segment a plurality of bone regions of the CT image sequence, the CT image sequence including one or more image slices;
b) in each image slice of the CT image sequence, determine a bone region with the maximum area among the plurality of bone regions as a spine region;
c) in each image slice of the CT image sequence, determine at least one anchor point in the top row of the spine region, and validate the at least one anchor point based on information associated with a position of the at least one anchor point in the image slice; and
d) calibrate the at least one anchor point based on the validation result,
wherein to validate the at least one anchor point, the system is further caused to:
determine a y coordinate of the at least one anchor point is greater than ¼ of the height of the image slice of the CT image sequence, and a shift of an x coordinate from the central position is less than ¼ of a width of the image slice of the CT image sequence; and
determine the at least one anchor point is valid.

9. The system of claim 8 , wherein to retrieve a CT image sequence and segment a plurality of bone regions of the CT image sequence,
the system is further caused to:
retrieve the CT image sequence;
select a preliminary spine region in each image slice of the CT image sequence; and
pre-process the preliminary spine region to segment the plurality of bone regions.

10. The system of claim 9 , wherein to select a preliminary spine region, the system is further caused to:
select and segment the CT image sequence based on information associated with a CT scan that generates the CT image sequence.

11. The system of claim 9 , wherein to pre-process the preliminary spine region, the system is further caused to:
preform a threshold segmentation on the preliminary spine region to identify the plurality of bone regions; and
eliminate deficiency and disconnections of the plurality of bone regions based on a morphology algorithm.

12. The system of claim 8 , wherein the validation result comprising that the at least one anchor point has an abnormal coordinate or lacks a coordinate, and to calibrate the at least one anchor point based on the validation result, the system is further caused to:
calibrate an image slice of the CT image sequence whose anchor point has an abnormal coordinate; or
calibrate an image slice of the CT image sequence whose anchor point lacks a coordinate.

13. The system of claim 12 , wherein to calibrate an image slice of the CT image sequence whose anchor point has an abnormal coordinate, the system is further caused to:
position the at least anchor point of at least one image slice of the CT image sequence sequentially;
define a sliding window;
position the at least one anchor point to be calibrated in the sliding window sequentially;
calculate a position mean value of the at least one anchor point in the sliding window; and
define a first Euclidean distance between the current anchor point and the position mean value as DM and a second Euclidean distance between the current anchor point and the previous anchor point as DP, wherein DM or DP is larger than a threshold TH, the coordinate of the current anchor point is substituted by a coordinate corresponding to the smaller of DM and DP.

14. The system of claim 12 , wherein to calibrate an image slice of the CT image sequence whose anchor point lacks coordinate, the system is further caused to:
fix the anchor point that lacks a coordinate based on an anchor point in an adjacent image slice of the CT image sequence.

15. A non-transitory computer readable medium comprising executable instructions that, when executed by a computing device, cause the computing device to:
a) retrieve a Computed Tomography (CT) image sequence and segment a plurality of bone regions of the CT image sequence, the CT imag

Assignees

• Shanghai United Imaging Healthcare Co Ltd

Inventors

• Wang Lilong
• Ma Jieyan
• Wang Jiyong
• Wang Ce
• Jin Lianghai
• Zhang Zhao
• Wang Fei
• Li Cheng
• Duan Qi
• Liu Wenqing
• Lyu Chuanfeng
• Qu Jie
• Luo Yong
• Pan Liangliang
• Wang Rui
• Cao Shili
• Mao Yufei
• Tan Bin
• Wang Wenhua
• Wang Shuai
• Zhang Fanfan

Classifications

• G06V10/50

  by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis · CPC title

• G06T7/187

  involving region growing; involving region merging; involving connected component labelling · CPC title

• G06T2207/30012

  Spine; Backbone · CPC title

• G06T2207/30008

  Bone · CPC title

• G06T2207/30081

  Prostate · CPC title