Rib suppression in radiographic images
A method for rib suppression in a chest x-ray image detects and labels one or more ribs in a region of interest and detects rib edges of the one or more detected ribs. Cross rib profiles are generated along the detected ribs. The original x-ray image is conditioned according to at least one of the detected rib edge and cross rib profiles. The conditioned x-ray image can be stored, displayed, or transmitted.
1. A method for rib suppression in a chest x-ray image, comprising:
detecting one or more ribs in a region of interest in the chest x-ray image;
applying a pixel classification to the one or more detected ribs in the region of interest;
using the pixel classification, labeling the one or more detected ribs in the region of interest;
detecting rib edges of the one or more detected ribs;
generating cross rib profiles along the detected rib edges;
conditioning the chest x-ray image according to at least one of the detected rib edges and cross rib profiles; and
storing, displaying, or transmitting the conditioned chest x-ray image.
2. The method of claim 1 wherein labeling the one or more ribs comprises segmenting a lung region of interest from the image.
3. The method of claim 1
wherein labeling the one or more ribs comprises extracting a medial axis from at least one of the one or more ribs.
4. The method of claim 1 further comprising adjusting segments of one or more of the detected rib edges to improve edge fitting.
5. The method of claim 1 further comprising scaling the image to a reduced resolution.
6. The method of claim 2 wherein labeling the one or more ribs in the image comprises classifying pixels within the lung region of interest as rib pixels or non-rib pixels.
7. The method of claim 1 wherein detecting the rib edges further comprises applying a model.
8. The method of claim 4 wherein adjusting segments comprises applying a growing or extending algorithm for rib edges.
9. The method of claim 3 further comprising obtaining a skeletal image comprising one or more of the medial axes.
10. The method of claim 1 further comprising smoothing the detected rib edges before generating cross rib profiles along the detected rib edges.
11. A method for rib suppression in a chest x-ray image, comprising:
segmenting one or more lung regions in the chest image;
determining one or more rib features for each pixel within the one or more segmented lung regions;
extracting a medial axis from at least one of the one or more determined rib features and iteratively segmenting one or more rib edges from the at least one or more rib features according to the extracted medial axis and according to an anatomical model;
forming a conditioned image by combining the one or more segmented rib edges with the original x-ray image; and
storing, displaying, or transmitting the conditioned image.
12. The method of claim 11 further comprising scaling the image to a reduced resolution.
13. The method of claim 11 wherein determining one or more rib features in the image comprises classification of pixels within the lung region of interest as rib pixels or non-rib pixels.
14. The method of claim 11 further comprising extending the one or more segmented rib edges by applying a growing algorithm.
15. The method of claim 11 wherein forming the conditioned image further comprises applying a filter to the image data.
16. The method of claim 11 wherein determining the one or more rib features further comprises grouping adjacent pixels associated with particular rib features.
17. A method for rib detection in a chest x-ray image, executed at least in part by a computer system, comprising:
detecting pixels for one or more ribs in a region of interest;
labeling one or more of the detected ribs as individual ribs by grouping the pixels to generate detected ribs and removing one or more false positive pixels from the detected pixels;
detecting one or more rib edges of the detected ribs by adjusting one or more edge segments; and
storing, displaying, or transmitting the detected results.
18. The method of claim 17 wherein detecting one or more rib edges further comprises detecting at least one medial axis for each rib and further comprises curve fitting.
19. The method of claim 17 wherein detecting one or more rib edges further comprises optimizing one or more line segments.
20. The method of claim 17 further comprising enhancing the one or more rib edges by applying a growth algorithm.