Image analysis – Applications – Biomedical applications
Reexamination Certificate
2000-08-14
2003-08-19
Werner, Brian (Department: 2621)
Image analysis
Applications
Biomedical applications
C382S173000
Reexamination Certificate
active
06608917
ABSTRACT:
BACKGROUND
1. Technical Field
This disclosure relates to radiographic imaging, and more particularly, to a system and method for detecting vertebra endplates in radiographic images.
2. Description of the Related Art
With the development of digital imaging technologies, the interest in using computers for assisting deformity analysis of scoliotic spines in radiography is increasing. Currently, most of these measurements are manually made. Manual measurement is not only time-consuming, but also subject to errors depending on the person's skill, experience and other human factors. Spine position information is obtained through X-rays of the spine. This information may be employed not only for deformity measurement of the spine itself but also as a reference position used for deformity quantification or measurements for many other anatomic landmarks.
Vertebra endplates are one important anatomical feature of the spine, which may be employed to quantitatively analyze and diagnose the spine. Vertebra endplate detection is difficult to automate for several reasons. For example, some vertebrae may appear dark, others bright, and some vertebra may appear obscured by other organs because of different imaging conditions and disease status. As a result, methods based on evidence provided by individual vertebrae may become unreliable.
Therefore, a need exists for an automatic system and method for detection of vertebra endplates for quantitative analysis and diagnosis. A further need exists for a system and method for determining the orientations of these endplates in radiographic images.
SUMMARY OF THE INVENTION
A method for detecting endplates of vertebra, in accordance with the present invention, provides an intensity curvature map of an image of a spine and filters the intensity curvature map in a direction relative to a spine axis. At a plurality of points of the spine axis, maximum and minimum scores are computed by summing curvatures of the intensity curvature map along line segments to determine if an endplate passes through the points. Endplate positions are found by comparing relative positions of plotted values corresponding to the maximum and minimum scores.
Another method for detecting endplates of vertebra, includes the steps of providing an image of a spine, providing an intensity curvature map of the image of a spine, filtering the intensity curvature map in a direction orthogonal to a spine axis, at a plurality of points of the spine axis, computing maximum and minimum scores by summing curvatures of the intensity curvature map and by summing directional intensity gradients along line segments to determine if an endplate passes through the points, performing an intensity gradient projection by determining an intensity gradient on the image of the spine in a direction orthogonal to line segments having maximum scores and determining endplate positions and orientations by employing relative positions of plotted values corresponding to the maximum and minimum scores and the intensity gradient projection.
In other methods, the step of performing an intensity gradient projection by determining an intensity gradient on the image of the spine in a direction orthogonal to line segments having maximum scores may be included. The step of performing an intensity gradient projection may include employing the intensity gradient projection to determine if the endplates are an upper endplate or a lower endplate. The step of computing maximum and minimum scores includes the steps of modeling an endplate by employing a line segment between spine boundaries, computing a co-linearity fit between the line segment and the intensity curvature map at a plurality of angular increments and computing scores as a sum of curvatures along each line segment. The method may include the step of confirming positions of the endplates by employing information about vertebra shapes.
In still other methods, the information about vertebra shapes may include, heights of vertebra and heights of inter-vertebral disks. The step of determining if endplate positions are found by comparing relative positions of peak values corresponding to the maximum and minimum scores, may include providing a score function which considers minimum and maximum scores and relative positions of the peak values corresponding to the maximum and minimum scores, and determining a maximum score in accordance with the score function to determine the endplates positions. The step of superimposing endplate images on the image of the spine may also be included. The methods of the present invention may be implemented by a program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform the method steps.
REFERENCES:
patent: 5982916 (1999-11-01), Kuhn
patent: 6249590 (2001-06-01), Young et al.
Tascini et al., “Automatic Quantitative Analysis of Lumbar Bone Radiographs,” Proceedings of the 1993 IEEE Nuclear Science Symposium & Medical Imaging Conference, vol. 3, pp. 1722-1726, San Fransisco, CA, Oct. 30-Nov. 6, 1993.*
Verdonck et al. “Computer Assisted Quatitative Analysis of Deformities of the Human Spine,” Proceedings of Medical Image Computing and Computer-Assisted Intervention, pp. 821-831, Cambridge, MA, Oct. 11-13, 1998.
Qian Jianzhong
Wei Guo-Qing
F. Chau & Associates LLP
Miller Ryan J.
Paschburg Donald B.
Siemens Corporate Research Inc.
Werner Brian
LandOfFree
Detection of vertebra endplates in digital radiography does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Detection of vertebra endplates in digital radiography, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Detection of vertebra endplates in digital radiography will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3093733