Automatic left ventricular function evaluation

1. A method for automatic left ventricular inner border detection, the method comprising:

performing segmentation on an echocardiogram by segmenting the echocardiogram simultaneously with pixel mapping using piece-wise histogram equalization, wherein the piecewise histogram equalization comprises transforming a pixel intensity histogram of a homogenous regions of the echocardiogram into an equalized histogram having an optimal uniform distribution of intensities, and assigning new intensity values to pixels of the homogenous regions of the echocardiogram based on said equalized histogram, thereby obtaining a multi-level image map having at least three uniformly distributed intensity levels, wherein a first intensity level represents blood, a second intensity represents noise and/or edges and a third intensity level represents myocardium;

converting the multi-level image map into a binary image by attributing pixels of one or more darker levels of the multilevel image map to the left ventricular cavity and pixels of one or more lighter levels of the image map to the myocardium;

identifying edges in the binary image;

applying a radial filter to the edges of the myocardium in the binary image to extract an approximate inner border of the left ventricular cavity and to separate the inner border from the outer border; and

performing shape modeling on the approximate inner border to complete representation of the left ventricular inner border, wherein the shape modeling is a polynomial shape modeling.

2. The method of claim 1 , further comprising, automatically detecting a region of interest in the echocardiogram, prior to performing the image mapping, wherein the region of interest encloses the left ventricle.

3. The method of claim 2 , wherein the automatic detection of the region of interest comprises:

producing an inverted binary version of the echocardiogram;

computing a column histogram of white pixels in the inverted binary version; and

defining an area delimited between two minimum points of the column histogram as the region of interest.

4. The method of claim 1 , further comprising defining the homogeneous regions of the echocardiogram based on an entropy threshold value.

5. The method of claim 4 , wherein the defining of the homogenous regions further comprises determining a central axis of the left ventricle and dividing the left ventricle appearing in the echocardiogram into a plurality of rectangular segments situated on lateral/anterior and septal/inferior sides of the central axis.

6. The ultrasonic imaging device according to claim 1 , wherein the polynomial shape modeling comprises a sixth-order polynomial interpolation performed on polar coordinates of the approximate inner border of the left ventricle.

7. The ultrasonic imaging device according to claim 6 , wherein the polynomial shape modeling further comprises a fourth-order polynomial interpolation on the approximate inner border at a septal/inferior side of the left ventricle, and second-order interpolation on the approximate inner border at a lateral/anterior side of the left ventricle.

8. The method of claim 1 , wherein the echocardiogram comprises an apical view echocardiogram.

9. The method according to claim 8 , wherein the apical view comprises a four-chamber apical view.

10. The method of claim 9 , wherein the multi-level image map comprises a 3-level image map, and wherein the one or more darker levels of the image map comprise two darker levels and the one or more lighter levels of the image map comprise one lighter level.

11. The method of claim 8 , wherein the apical view comprises a two-chamber apical view.

12. The method of claim 11 , wherein the multi-level image map comprises a 3 to 5-level image map.

13. An ultrasonic imaging device, comprising:

an ultrasonic probe configured to acquire an echocardiogram; and

a processing unit connected to said probe, said processing unit configured to:

performing segmentation on an echocardiogram by segmenting the echocardiogram simultaneously with pixel mapping using piece-wise histogram equalization, wherein the piecewise histogram equalization comprises transforming a pixel intensity histogram of a homogenous regions of the echocardiogram into an equalized histogram having an optimal uniform distribution of intensities, and assigning new intensity values to pixels of the homogenous regions of the echocardiogram based on said equalized histogram, thereby obtaining a multi-level image map having at least three uniformly distributed intensity levels, wherein a first intensity level represents blood, a second intensity represents noise and/or edges and a third intensity level represents myocardium;

converting the multi-level image map into a binary image by attributing pixels of one or more darker levels of the multilevel image map to the left ventricular cavity and pixels of one or more lighter levels of the image map to the myocardium;

identifying edges in the binary image;

applying a radial filter to the edges of the myocardium in the binary image to extract an approximate inner border of the left ventricular cavity and to separate the inner border from the outer border; and

performing shape modeling on the approximate inner border to complete representation of the left ventricular inner border, wherein the shape modeling is a polynomial shape modeling.

14. The ultrasonic imaging device of claim 13 , further comprising defining homogeneous regions in the echocardiogram, based on an entropy threshold value.

15. The ultrasonic imaging device of claim 13 , wherein the echocardiogram comprises an apical view echocardiogram.

16. The ultrasonic imaging device of claim 15 , wherein the apical view comprises a four-chamber apical view.

17. The ultrasonic imaging device of claim 13 , wherein said processing unit is further configured to automatically detect a region of interest in the echocardiogram prior to performing the image mapping, wherein the region of interest encloses the left ventricle.

18. The ultrasonic imaging device of claim 17 , wherein the automatic detection of the region of interest comprises:

producing an inverted binary version of the echocardiogram;

computing a column histogram of white pixels in the inverted binary version; and

defining an area delimited between two minimum points of the column histogram as the region of interest.