System and method for detecting a three dimensional flexible tube in an object
View Patent ↗The present invention is directed to a system and method for populating a database with a set of image sequences of an object. The database is used to detect a tubular structure in the object. A set of images of objects are received in which each image is annotated to show a tubular structure. For each given image, a Probabilistic Boosting Tree (PBT) is used to detect three dimensional (3D) circles. Short tubes are constructed from pairs of approximately aligned 3D circles. A discriminative joint shape and appearance model is used to classify each short tube. A long flexible tube is formed by connecting all of the short tubes. A tubular structure model that comprises a start point, end point and the long flexible tube is identified. The tubular structure model is stored in the database.
1. A method for populating a database with a set of image sequences of an object, the database being used to detect a tubular structure in the object, the method comprising the steps of:
receiving a set of images of objects, each image being annotated to show a tubular structure;
for each given image:
using a Probabilistic Boosting Tree (PBT) to detect three dimensional (3D) circles;
constructing short tubes from pairs of approximately aligned 3D circles;
using a discriminative joint shape and appearance model to classify each short tube;
constructing a long flexible tube by connecting all of the short tubes;
identifying a tubular structure model that comprises a start point, end point and the long flexible tube; and
storing the tubular structure model in the database.
2. The method of claim 1 wherein 3D circles are selected based upon a voting strategy.
3. The method of claim 1 wherein the step of constructing short tubes from pairs of approximately aligned 3D circles further comprises the steps of:
identifying a predetermined number of candidate tubes on each side of the detected 3D circle;
pairing each candidate tube with the detected 3D circle;
determining a likelihood probability for each pair; and
maintaining those pairs whose probability is above a predetermined value.
4. The method of claim 1 wherein the object is an anatomical structure.
5. The method of claim 4 wherein the anatomical structure is a colon.
6. The method of claim 5 wherein the tubular structure is a rectal tube.
7. A system for detecting a tubular structure in an object captured in an image sequence, the system comprising:
a database that stores a set of image sequences of objects, each image being annotated to show a tubular structure;
a processor that receives an input image sequence of an object, the processor performing the following steps on each image in the image sequence:
using a Probabilistic Boosting Tree (PBT) to detect 3D circles in the image;
constructing short tubes from the 3D circles;
using a discriminative joint shape and appearance model to classify each short tube;
constructing a long flexible tube by connecting all of the short tubes;
identifying a tubular structure model that comprises a start point, end point and the long flexible tube;
storing the tubular structure model in the database; and
a display to display the tubular structure in the object.
8. The system of claim 7 wherein 3D circles are selected based upon a voting strategy.
9. The system of claim 7 wherein the step of constructing short tubes from pairs of approximately aligned 3D circles further comprises the steps of:
identifying a predetermined number of candidate tubes on each side of the detected 3D circle;
pairing each candidate tube with the detected 3D circle;
determining a likelihood probability for each pair; and
maintaining those pairs whose probability is above a predetermined value.
10. The system of claim 7 wherein the object is an anatomical structure.
11. The system of claim 10 wherein the anatomical structure is a colon.
12. The system of claim 11 wherein the tubular structure is a rectal tube.