Granted Patent B2

US 7,519,207 · App. 10/993,055 · Granted Apr 14, 2009

Abstract

Claims (68)

1. A method of detecting an orientation of a digital radiograph, comprising the steps of:

accessing the radiograph;

locating a region of interest in the radiograph; and

using the located region of interest and a pre-trained classifier to detect the orientation of the radiograph;

wherein the method further comprise the step of, prior to locating a region of interest in the radiograph, preprocessing the radiograph;

wherein the step of preprocessing the radiograph comprises the steps of:

sub-sampling the radiograph;

segmenting the radiograph into foreground, background, and anatomy regions;

removing the foreground region and the background region from the radiograph;

and normalizing the intensity of the anatomy region of the radiograph.

2. The method of claim 1 , wherein the step of locating the region of interest comprises the steps of:

detecting a medial axis of an anatomy region of the radiograph;

locating a center of the region of interest;

determining the size and shape of the region of interest; and

extracting the region of interest.

3. The method of claim 1 , wherein the step of using the located region of interest and a classifier to detect the orientation of the radiograph includes the steps of:

extracting features of the radiograph; and

classifying the extracted features.

4. A method of detecting an orientation of a digital radiograph, comprising the steps of:

accessing the radiograph;

locating a region of interest in the radiograph; and

using the located region of interest and a pre-trained classifier to detect the orientation of the radiograph;

wherein the step of locating the region of interesting comprises the steps of:

detecting a medial axis of an anatomy region of the radiograph;

locating a center of the region of interest;

determining the size and shape of the region of interest; and

extracting the region of interest;

wherein the step of detecting the medial axis of the anatomy region comprises of the steps of:

detecting a contour of the anatomy region;

computing a distance map of the contour; and

detecting a ridge in the distance map as the medial axis.

5. A method of detecting an orientation of a digital radiograph, comprising the steps of:

accessing the radiograph;

locating a region of interest in the radiograph; and

using the located region of interest and a classifier to detect the orientation of the radiograph;

wherein the step of using the located region of interest and a classifier to detect the orientation of the radiograph comprises the steps of:

extracting a set of orientation related features from the region of interest;

identifying an orientation of the radiograph using the classifier wherein the classifier is a pre-trained classifier;

rotating the radiograph if the orientation is incorrect; and

repeating the step of identifying and rotating until a correct orientation is identified.

6. The method of claim 5 , wherein the pre-trained classifier is obtained by the steps of:

collecting a pre-determined number of training radiograph images, each having a known orientation;

locating a region of interest for each training image;

computing a set of features from the region of interest of each training image;

associating a target output specifying the orientation of each training image;

computing a transformation for the feature set; and

training a classifier with the transformed feature set and target output to generate the pre-trained classifier.

7. Method of claim 6 , wherein the step of computing a set of features comprises of the steps of:

locating a plurality of peripheral sub-blocks of the region of interest; and

computing a set of features using only the located plurality of peripheral sub-blocks.

8. A method of detecting and orienting a digital radiograph, comprising the steps of;

accessing the radiograph;

preprocessing the radiograph by;

(a) sub-sampling the radiograph;

(b) segmenting the radiograph into foreground, background, and anatomy regions;

(c) removing the foreground region and the background region from the radiograph; and

(d) normalizing the intensity of the anatomy region of the radiograph;

locating a region of interest in the radiograph;

using the located region of interest and a pre-trained classifier to detect the orientation of the radiograph; and

reorienting the radiograph to a preferred orientation for display on a display;

wherein the step normalizing the intensity of the anatomy region of the radiograph comprises the steps of:

detecting a medial axis of the anatomy region;

locating a center of the region of interest;

determining a size and shape of the region of interest; and

extracting the region of interest;

wherein the step of detecting the medial axis of the anatomy region comprises of the steps of:

detecting a contour of the anatomy region;

computing a distance map of the contour; and detecting a ridge in the distance map as the medial axis.

Assignments (15)