Weighted inverse topography method for digital x-ray image data processing

• US 6,381,351 B1
• Filed: 11/24/1999
• Issued: 04/30/2002
• Est. Priority Date: 11/24/1999
• Status: Expired due to Fees

• Current Assignee: Direct Radiography Corp. (Hologic Incorporated)
• Original Assignee: Direct Radiography Corp. (Hologic Incorporated)
• Inventors: Powell, Gregory F.
• Primary Examiner(s): Mancuso, Joseph
• Assistant Examiner(s): TABATABAI, ABOLFAZL
• Application Number:
  US09/449,073
• Time in Patent Office: 888 Days
• Field of Search: 382/128, 382/129, 382/130, 382/131, 382/132, 382/264, 250/455
• US Class Current: 382/131
• CPC Class Codes: G01N 23/04 and forming images of the material
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction

Abstract

A method for displaying on a display medium an image corresponding to detected exposures of a plurality of pixels converted to a matrix having n rows and m columns of digital values. The method comprises creating a weighted, smoothed, subtracted matrix, adding the weighted, smoothed, subtracted matrix to the starting matrix to produce an enhanced matrix; and then using the digital values from the enhanced matrix to display an image. Preferably, a sampled submatrix of digital values having n/I rows and m/I columns is created from the starting matrix from which is compiled a weighted, smoothed, subtracted submatrix that is then extrapolated to create the weighted, smoothed, subtracted matrix. A program storage device for carrying out the method, a digital x-ray image data capture system, and a method of using the system of this invention are also claimed.

Claims

1. A method for displaying on a display medium an image corresponding to detected exposures to radiation of a plurality of sensors in an array, the detected exposures converted to a starting matrix having n rows and m columns of digital values, each digital value representing an optical density, the method comprising the steps of:

(a) creating a weighted, smoothed, subtracted matrix by;

(a1) sampling the starting matrix to generate a sampled submatrix having n/I rows and m/I′

columns of submatrix digital values, wherein I and I′

are integers greater than 1;

(a2) creating a subtracted submatrix by subtracting each submatrix digital value in the sampled submatrix or in a first intervening submatrix from a reference digital value greater than each submatrix digital value;

(a3) expanding the subtracted submatrix or a second intervening submatrix by extrapolation, interpolation, or both, to create an expanded matrix having n rows and m columns;

(a4) performing a smoothing step between step (a1) and step (a3); and

(a5) performing a weighting step sometime after step (a1);

(b) adding the weighted, smoothed, subtracted matrix to the starting matrix to produce an enhanced matrix;

(c) using the digital values from the enhanced matrix to display an image.

2. The method according to claim 1 wherein step (a4) comprises applying a smoothing filter to the rows of the sampled submatrix or the subtracted submatrix to create a row-smoothed submatrix, applying a smoothing filter to the columns of the sampled submatrix or the subtracted submatrix to create a column-smoothed submatrix, and then adding the row-smoothed submatrix and the column-smoothed submatrix.

3. The method according to claim 1 wherein I is equal to I′

.

4. The method according to claim 1 further comprising expanding the subtracted submatrix or second intervening submatrix in step (a3) by inserting interpolated digital values between each of the digital values in the n/I rows and m/I′

columns.

5. The method according to claim 1 wherein the interpolation in step (a3) comprises bilinear interpolation.

6. The method according to claim 1 further comprising in step (c) normalizing the enhanced matrix, mapping the normalized, enhanced matrix onto a look up table to create a matrix of digital display values, and sending the matrix of digital display values to a display medium for displaying an image.

7. The method of claim 1 further comprising reducing the dynamic range the digital values in one of:

the sampled submatrix in step (a1), the enhanced matrix in step (b), or both;

wherein reducing the dynamic range comprises reducing the dynamic range in one of;

an upper range of the data values, a lower range of the data values, or both.

8. The method according to claim 1 wherein step (a5) is performed after step (a3).

9. A method for displaying on a display medium an image corresponding to detected exposures to radiation of a plurality of sensors in an array, the detected exposures converted to a starting matrix (MATRIX_0) having n rows and m columns of digital values, each digital value representing an optical density, the method comprising the steps of:

(a) creating a weighted, smoothed, subtracted matrix by the steps of;

(i) sampling starting matrix MATRIX_0 to create a sampled submatrix having n/I rows and m/I′

columns, wherein I and I′

are integers, and then creating a maximum-limited, sampled submatrix by replacing any digital values in the sampled submatrix greater than a predetermined maximum value with the predetermined maximum value;

(ii) compiling a row-smoothed submatrix (MATRIX_1) by applying a linear smoothing filter to the digital values along the rows of the maximum-limited, sampled submatrix;

(iii) compiling a column-smoothed submatrix (MATRIX_2), by applying a linear smoothing filter to the digital values along the columns in the maximum-limited, sampled submatrix;

(iv) creating a row-column-smoothed submatrix (MATRIX_3) by adding together MATRIX_1 and MATRIX_2;

(v) creating a subtracted, smoothed submatrix (MATRIX_4) by subtracting each value in MATRIX_3 from a fixed value larger than each value in MATRIX_3; and

(vi) multiplying each value in MATRIX_4 by a weighing factor to created a weighted, smoothed, subtracted submatrix; and

(vii) expanding the weighted, smoothed, subtracted submatrix by extrapolation, interepolation, or both, to create the weighted, smoothed, subtracted matrix;

(b) adding the weighted, smoothed, subtracted matrix to the starting matrix to produce an enhanced matrix; and

(c) using the digital values from the enhanced matrix to display an image.

10. The method of claim 9 further comprising selecting an anchor point, in step (a)(i) setting the predetermined maximum value equal to a first constant k₁ times the anchor point, and in step (a)(v) setting the fixed value equal to a second constant k₂ times the anchor point.

11. The method of claim 10 further comprising developing a histogram corresponding to a frequency of occurrence for each digital value in the matrix of digital values, identifying a range of useful data within the histogram, calculating a statistical mode of the useful data within the histogram, and selecting as the anchor point the lesser of:

the statistical mode or a pre-selected safety point.

12. The method of claim 10 further comprising developing a histogram corresponding to a frequency of occurrence for each digital value in the matrix of digital values, identifying a range of useful data within the histogram, calculating a statistical mode of the useful data within the histogram, and selecting the statistical mode as the anchor point.

13. The method of claim 12 further comprising reducing the histogram prior to identifying the range of useful data within the histogram.

14. The method of claim 13 further comprising smoothing the histogram prior to calculating the statistical mode.

15. A method for displaying on a display medium an image corresponding to detected exposures of a plurality of sensors, the method comprising the steps of:

a) creating and storing in a data bank a starting matrix (MATRIX_0) of digital values each representing an optical density corresponding to the detected exposures, the starting matrix having n rows and m columns;

b) developing a histogram corresponding to a frequency of occurrence of each digital value and calculating a statistical mode (MODE) of at least a portion of data within the histogram;

c) creating a sampled submatrix of digital values having n/I rows and m/I columns by selecting every I^th digital value in the starting matrix, I being an integer;

d) creating a maximum-limited, sampled submatrix having n/I rows and m/I columns by replacing any digital values in the sampled submatrix greater than a predetermined maximum value with the predetermined maximum value, the predetermined maximum value equal to a constant k₁ times the statistical mode (MODE);

e) compiling a row-smoothed submatrix (MATRIX_1) by applying a linear smoothing filter to the digital values along the rows of the maximum-limited, sampled submatrix;

f) compiling a column-smoothed submatrix (MATRIX_2), by applying a linear smoothing filter to the digital values along the columns in the maximum-limited, sampled submatrix;

g) creating a row-column-smoothed submatrix (MATRIX_3) by adding together the row-smoothed submatrix (MATRIX_1) and the column-smoothed submatrix (MATRIX_2);

h) creating an subtracted, smoothed submatrix (MATRIX_4) by subtracting each value in the row-column-smoothed submatrix (MATRIX_3) from a fixed value equal to a constant k₂ times the statistical mode (MODE); and

i) multiplying each value. in the subtracted, smoothed submatrix (MATRIX_4) by a weighing factor w to create a weighted, smoothed, subtracted submatrix (MATRIX_5) having n/I rows and m/I columns;

j) creating a weighted, smoothed, subtracted matrix (MATRIX_6) having n rows and m columns by bilinear interpolation of the digital values in the n/I rows and m/I columns of the weighted, smoothed, subtracted submatrix;

k) adding the weighted, smoothed, subtracted matrix (MATRIX_6) to the starting matrix (MATRIX_0) to create an enhanced matrix (MATRIX_7); and

l) normalizing the enhanced matrix (MATRIX_7) and mapping the normalized, enhanced matrix onto a look up table comprising a gray scale transfer function to create a matrix of digital display values, and sending the digital display values to the display medium.

16. The method of claim 15 wherein the starting matrix (MATRIX_0) comprises a plurality of 14-bit digital values and step (a) comprises converting each 14-bit digital value in the starting matrix to a 12-bit, logarithmic digital value, developing the histogram corresponding to the frequency of occurrence of each 12-bit, logarithmic digital value, reducing the histogram by partitioning the histogram into bins of sequential 12-bit, logarithmic digital values, identifying a range of useful data within the histogram, smoothing the histogram, calculating a 12-bit, logarithmic, binned statistical mode of the useful data within the histogram, and converting the 12-bit, logarithmic, binned statistical mode to a 14-bit, non-logarithmic, non-binned converted value to be used as the statistical mode (MODE).

17. The method of claim 15 wherein k₁ is equal to about 5 and k₂ is equal to about 2.

18. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform a set of method steps for displaying on a display medium an image corresponding to detected exposures to radiation of a plurality of sensors, the detected exposures converted to a starting matrix having n rows and m columns of digital values stored in the program storage device, each digital value representing an optical density, the method steps comprising:

(a) creating a weighted, smoothed, subtracted matrix by;

(a1) sampling the starting matrix to generate a sampled submatrix having n/I rows and m/I′

columns of submatrix digital values, wherein I and I′

are integers greater than 1;

(a2) creating a subtracted submatrix by subtracting each submatrix digital value in the sampled submatrix or a first intervening submatrix from a reference digital value greater than each submatrix digital value;

(a3) expanding the subtracted submatrix or a second intervening submatrix by extrapolation, interpolation, or both, to create an expanded matrix having n rows and m columns;

(a4) performing a smoothing operation between step (a1) and step (a3); and

(a5) performing a weighting operation sometime after step (a1), optionally after step (a3);

(b) adding the weighted, smoothed, subtracted matrix to the starting matrix to produce an enhanced matrix;

(c) using the digital values from the enhanced matrix to display an image.

19. A digital x-ray image data capture system comprising:

the program storage device of claim 18;

a source of penetrative radiation for emitting an unmodulated radiation beam along a path;

a plurality of sensors in an array of n rows and m columns positioned in the beam path for receiving and detecting exposure to at least a modulated radiation beam derived from said unmodulated radiation beam, each sensor adapted to produce a sensor output proportional to the detected exposure;

an analog-to-digital converter adapted to receive the sensor output, convert the sensor output to a digital value, and transmit the digital value to the program storage device.

20. The digital x-ray image data capture system of claim 19 further comprising a portion of anatomy placed in the path of said unmodulated radiation beam between said source and said array of sensors, said anatomy capable of modulating said radiation beam to produce said modulated beam.

21. A method for displaying on a display medium a radiographic image of a portion of anatomy corresponding to detected exposures to radiation of a plurality of sensors in an array, the method comprising the steps of:

a) providing a digital x-ray image data capture system comprising;

a source of penetrative radiation for emitting an unmodulated radiation beam along a path;

the plurality of sensors in the array, the array having n rows and m columns, the plurality of sensors positioned in the beam path;

an analog-to-digital converter for receiving the sensor outputs, converting the outputs to digital values, and transmitting the digital values;

a program storage device tangibly embodying a program of instructions and for receiving and storing the digital values;

a program execution machine for reading the program storage device and executing the program of instructions;

b) placing the portion of anatomy in the path of the unmodulated radiation beam between the source and the array of sensors;

c) modulating the unmodulated radiation beam into a modulated beam as the beam penetrates the portion of anatomy;

d) receiving and detecting exposure to the modulated beam with the sensors and producing a sensor output proportional to the detected exposure;

e) receiving the sensor output in the analog-to-digital converter, converting the sensor outputs to the digital values, each digital value representing an optical density, and transmitting the digital values to the program storage device to be stored in a starting matrix having n rows and m columns; and

f) the program execution machine modifying the digital values by;

(i) creating a weighted, smoothed, subtracted matrix from the starting matrix,

(ii) adding the weighted, smoothed, subtracted matrix to the starting matrix to produce an enhanced matrix; and

(iii) using the digital values from the enhanced matrix to display the radiographic image of the portion of anatomy.

22. A method for displaying on a display medium an image corresponding to detected exposures to radiation of a plurality of sensors in an array, the detected exposures converted to a starting matrix having n rows and m columns of digital values, each digital value representing an optical density, the method comprising the steps of:

(a) creating a weighted, smoothed, subtracted matrix by;

(a1) sampling the starting matrix to generate a sampled submatrix having n/I rows and m/I′

columns of sampled digital values, wherein I and I′

are integers;

(a2) filtering said sampled submatrix with a smoothing filter to obtain a smoothed submatrix having filtered digital values;

(a3) creating a smoothed, subtracted submatrix by subtracting each of said submatrix digital values from a reference digital value; and

(a4) expanding the smoothed, subtracted submatrix by extrapolating the subtracted digital values to create a smoothed, subtracted matrix having n rows and m columns;

(b) adding the weighted, smoothed, subtracted matrix to the starting matrix to produce an enhanced matrix;

(c) using the digital values from the enhanced matrix to display an image; and

(d) reducing the dynamic range of the digital values in one of;

the sampled submatrix in step (a1), the enhanced matrix in step (b), or both;

wherein reducing the dynamic range comprises reducing the dynamic range in one of;

an upper range of the data values, a lower range of the data values, or both.