System and methods for processing images to measure multi-leaf collimator, collimator jaw, and collimator performance utilizing pre-entered characteristics

Patent metadata

Field	Value
Publication number	US-9480860-B2
Application number	US-201314040202-A
Country	US
Kind code	B2
Filing date	Sep 27, 2013
Priority date	Sep 27, 2013
Publication date	Nov 1, 2016
Grant date	Nov 1, 2016

Abstract

Systems and methods are proposed for accurate and efficient automatic measurement of jaw and leaf positioning in multi-leaf collimator imaging systems. Specifically, the method enables the automated and objective processing of images to determine characteristics of collimator jaws and MLC leaves. These novel techniques enable verification of collimator component positioning to ensure accurate beam modulation for radiation therapy procedures.

First claim

What is claimed is:

1. A method for image processing for collimator positioning, the method comprising:
a. receiving image data corresponding to an image generated by an imaging system comprising a multi-leaf collimator (MLC), a detector, and a radiation source;
b. calculating an expected position of an edge of a first leaf of the MLC based on pre-entered characteristics corresponding to the imaging system and the image data;
c. deriving a minimum position and a maximum position based on the expected position for the edge of the first leaf, the minimum and maximum positions corresponding to a width of the first leaf and along an axis corresponding to an orientation of the MLC;
d. verifying an availability of the minimum and maximum positions;
e. interpolating a minimum pixel value based on the minimum position and a maximum pixel value based on the maximum position;
f. generating a target pixel value based on an average pixel value of the minimum pixel value and the maximum pixel value; and
g. determining a target position based on the target pixel value.

2. The method according to claim 1 , further comprising:
h. determining a difference value between the target position and the expected position;
i. comparing the difference value to a threshold value; and
j. designating the target position as a location of the edge of the first leaf when the difference value is less than the threshold value.

3. The method according to claim 2 , further comprising:
repeating steps b)-j) for a plurality of iterations until the difference value is less than the threshold value,
wherein for each iteration of the plurality of iterations, the target position is used as the expected position for the subsequent iteration.

4. The method according to claim 2 , further comprising:
converting the target position to a plurality of units of length at an isoplane using a pixel pitch of the image as a scale factor.

5. The method according to claim 1 , wherein the generating a target pixel value comprises:
plotting a curve in a coordinate space corresponding to a pixel value and a position for a plurality of points in the image between the minimum and maximum positions.

6. The method according to claim 5 , further comprising performing a bisecting method on the curve to determine the target position.

7. The method according to claim 1 , wherein the image data comprises at least one image datum from a group consisting of:
an MLC image origin position;
an MLC center of rotation;
an estimated location of an isocenter;
a collimator angle;
a pixel pitch;
an expected leaf width;
an expected leaf end position.

8. The method according to claim 2 , further comprising locating a plurality of leaf side edges corresponding to the first leaf.

9. The method according to claim 8 , wherein locating the plurality of leaf side edges comprises determining a line segment bounded by the minimum position and the maximum position based on a measured end position of the first leaf and a measured leaf and end position of a second leaf, the second leaf neighboring the first leaf.

10. The method according to claim 8 , further comprising calculating a leaf centerline offset, the leaf centerline offset comprising an average value of the plurality of leaf side edges.

11. The method according to claim 10 , further comprising:
determining a collimator angle based on the leaf centerline offset;
determining a pixel pitch based on the leaf centerline offset; and
determining an MLC general offset based on the leaf centerline offset.

12. The method according to claim 11 , further comprising:
k. calculating a rotation error based on the leaf centerline offset;
l. calculating a collimator angle error value based on an average of a plurality of rotation errors;
m. updating the collimator angle with the collimator angle error to create an updated collimator angle;
n. reprocessing the image using the updated collimator angle to define an MLC coordinate frame.

13. The method according to claim 12 , further comprising:
o. for each bank of a plurality of banks of leaves comprising the MLC, calculating a linear relation between the leaf centerline offset and an expected leaf centerline offset;
p. calculating a general offset error based on the linear relation;
q. updating the MLC general offset with the general offset error to create an updated MLC general offset;
r. calculating a magnification factor error based on the linear relation;
s. updating the pixel pitch with the magnification factor error to create an updated pixel pitch; and
t. reprocessing the image using the updated MLC general offset and pixel pitch to define the MLC coordinate frame.

14. The method according to claim 13 , further comprising:
u. comparing the updated collimator angle to an collimator angle threshold value;
v. comparing the updated MLC general offset to an MLC general offset threshold value;
w. comparing the updated pixel pitch to a pixel pitch threshold value; and
x. repeating steps k.-w. until an updated collimator angle error is less than a collimator angle error threshold value, an updated MLC general offset error is less than an MLC general offset error threshold value, and an updated pixel pitch error is less than a pixel pitch error threshold value.

15. A computer readable medium containing program instructions embodied therein for causing a computer system to process a generated image to determine a plurality of measurements of an imaging system, the program instructions comprising:
instructions to receive image data corresponding to an image generated by an imaging system comprising a multi-leaf collimator (MLC), a detector, and a radiation source;
instructions to calculate an expected position of an edge of a first leaf of the MLC based on pre-entered characteristics corresponding to the imaging system and the image data;
instructions to derive a minimum position and a maximum position based on the expected position for the edge of the first leaf, the minimum and maximum positions corresponding to a width of the first leaf and along an axis corresponding to an orientation of the MLC;
instructions to verify an availability of the minimum and maximum positions;
instructions to interpolate a minimum pixel value based on the minimum position and a maximum pixel value based on the maximum position;
instructions to generate a target pixel value based on an average pixel value of the minimum pixel value and the maximum pixel value; and
instructions to determine a target position based on the target pixel value.

Assignees

• Varian Med Sys Inc

Inventors

• Gaudio Stephen

Classifications

• A61N5/1081

  Rotating beam systems with a specific mechanical construction, e.g. gantries · CPC title

• A61N5/1045 Primary

  using a multi-leaf collimator, e.g. for intensity modulated radiation therapy or IMRT · CPC title

• A61N2005/1095

  Elements inserted into the radiation path within the system, e.g. filters or wedges · CPC title

• A61N5/1075

  for testing, calibrating, or quality assurance of the radiation treatment apparatus · CPC title