Methods and systems for automated detection and analysis of lesion on magnetic resonance images
View Patent ↗A method for detecting and analyzing candidate lesions in a magnetic resonance image of a breast. The method includes the steps of: accessing a plurality of temporal magnetic resonance images of the breast; identifying candidate lesions by performing a temporal pattern analysis of the plurality of images to produce temporal features based on an uptake phase and a washout phase; performing a morphological operation on the candidate lesions to produce morphological features; and classifying the candidate lesions using the morphological features and temporal features to produce classified candidate lesions.
1. A method for detecting and
analyzing candidate lesions in a plurality of digital magnetic resonance images of a breast, the plurality of digital magnetic resonance images including both a pre-contrast image and a post-contrast plurality of temporal images, the method comprising using a computer to perform steps of:
accessing the pre-contrast image and the post-contrast a plurality of temporal images;
identifying candidate lesions by performing a pixel by pixel temporal pattern analysis of the pre-contrast image and the post-contrast plurality of temporal images to produce temporal patterns based on a plurality of categories of an uptake rate of a contrast agent in an uptake phase and a plurality of categories of washout rate of the contrast agent in a washout phase; the identifying step includes steps of:
calculating a relative enhancement for each post-contrast temporal image,
applying an uptake threshold to each relative enhancement to identify suspect regions
determining the candidate lesions from the suspect regions
determining a curve type for each candidate lesion using the relative enhancement in the uptake phase and an enhancement ratio in the washout phase, wherein the curve type are predefined combinations of an uptake rate and the plurality of categories of washout rate; and
classifying the candidate lesions based on temporal patterns to produce classified candidate lesions.
2. The method of claim 1 further comprising a step of:
classifying each candidate lesion into one of nine categories based on the curve type.
3. The method of claim 1 , further comprising a step of:
displaying at least some of the classified candidate lesions on a display.
4. The method of claim 1 , further comprising a step of:
generating a color map representative of the temporal patterns.
5. The method of claim 1 , further comprising a step of:
generating a color map representative of the candidate lesions.
6. The method of claim 1 , further comprising a step of:
prior to identifying the candidate lesions, registering the plurality of temporal images.
7. The method of claim 1 , further comprising a step of:
determining a likelihood of the candidate lesion being an actual abnormality based on the output of the classifying step.
8. A method for detecting and analyzing candidate lesions in a plurality of digital magnetic resonance images of a breast, the plurality of digital magnetic resonance images including both a pre-contrast image and a post-contrast plurality of temporal images, the method comprising using a computer to perform steps of:
accessing the pre-contrast image and the post-contrast plurality of temporal images,
identifying candidate lesions by performing a pixel by pixel temporal pattern analysis of the pre-contrast image and the post-contrast plurality of temporal images to produce temporal patterns based on a plurality of categories of uptake rate of a contrast agent in an uptake phase and a plurality of categories of washout rate of the contrast agent in a washout phase, the identifying step including steps of:
calculating a relative enhancement for each post-contrast temporal image;
applying an uptake threshold to each relative enhancement to identify suspect regions;
determining the candidate lesions from the suspect regions; and
determining a curve type for each candidate lesion using the relative enhancement in the uptake phase and an enhancement ratio in the washout phase, wherein the curve types are predefined combinations of an uptake rate and a washout rate from the plurality of categories of uptake rate and the plurality of categories of washout rate;
classifying the candidate lesions using the temporal patterns to produce classified candidate lesions; and
generating a color map representative of the classified candidate lesions.
9. The method of claim 8 , further comprising steps of:
classifying each candidate lesion into one of nine categories based on the curve type; and
generating a color map representative of the curve type.
10. The method of claim 8 , further comprising steps of:
classifying each classified candidate lesion into one of nine categories based on the curve type; and
generating a color map representative of the curve type.
11. A method according to claim 1 , further comprising steps of:
performing a morphological operation on the candidate lesions to produce morphological features; and
further classifying the candidate lesions using the morphological features.
12. A method according to claim 1 , wherein the step of determining the candidate lesions comprises calculating an average enhancement and an absolute enhancement of the suspect lesions.
13. A method according to claim 8 , further comprising steps of:
performing a morphological operation on the candidate lesions to produce morphological features; and
further classifying the candidate lesions using the morphological features.
14. A method according to claim 8 , wherein the step of determining the candidate lesions comprises calculating an average enhancement and an absolute enhancement of the suspect lesions.
15. The method of claim 8 , further comprising a step of:
prior to identifying the candidate lesions, registering the plurality of temporal images.