Automated septal defect detection in cardiac computed tomography images
Automated detection of septal defects in cardiac computed tomography images is provided. In various embodiments, a blood pool is extracted from a CTA image of a heart. The blood pool is determined to be a left pool. A septal curve of the left pool is extracted. A plurality of features of the septal curve is determined. The heart is classified as having an atrial septal defect, ventricular septal defect, or no defect based on the plurality of features.
1. A method comprising:
extracting a blood pool from a CTA image of a heart;
determining that the blood pool is a left pool based on an elongation of the blood pool;
extracting a septal curve of the left pool, wherein extracting the septal curve comprises determining a major principal axis of the left pool;
determining a plurality of features of the septal curve;
classifying the heart as having an atrial septal defect, ventricular septal defect, or no defect based on the plurality of features, wherein classifying the heart comprises applying a support vector machine or random forest.
2. The method of claim 1 , wherein extracting the blood pool comprises thresholding.
3. The method of claim 2 , wherein the thresholding comprises applying k-means or Otsu methods.
4. The method of claim 2 , wherein thresholding comprises clustering image intensities into a predetermined number of classes.
5. The method of claim 1 , further comprising:
extracting a 2D CTA image from a 3D CTA study.
6. A method comprising:
extracting a blood pool from a CTA image of a heart;
determining that the blood pool is a combined pool based on an elongation of the blood pool;
determining a plurality of features of the combined pool, wherein determining the plurality of features comprises determining a septal point by locating a deepest convexity defect of the combined pool;
classifying the heart as having an atrial septal defect or ventricular septal defect based on the plurality of features.
7. The method of claim 6 , wherein extracting the blood pool comprises thresholding.
8. The method of claim 7 , wherein the thresholding comprises applying k-means or Otsu methods.
9. The method of claim 7 , wherein thresholding comprises clustering image intensities into a predetermined number of classes.
10. The method of claim 6 , wherein classifying the heart comprises applying a support vector machine or random forest.
11. The method of claim 6 , further comprising:
extracting a 2D CTA image from a 3D CTA study.
12. A computer program product for septal defect detection, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:
extracting a blood pool from a CTA image of a heart;
determining that the blood pool is a left pool based on an elongation of the blood pool;
extracting a septal curve of the left pool, wherein extracting the septal curve comprises determining a major principal axis of the left pool;
determining a plurality of features of the septal curve;
classifying the heart as having an atrial septal defect, ventricular septal defect, or no defect based on the plurality of features, wherein classifying the heart comprises applying a support vector machine or random forest.
13. The computer program product of claim 12 , wherein extracting the blood pool comprises thresholding.
14. The computer program product of claim 13 , wherein the thresholding comprises applying k-means or Otsu methods.
15. The computer program product of claim 13 , wherein thresholding comprises clustering image intensities into a predetermined number of classes.
16. The computer program product of claim 12 , the method further comprising:
extracting a 2D CTA image from a 3D CTA study.