Abnormality detection in medical images
View Patent ↗A system for the detection of abnormalities in a medical image of a subject. The system includes an examination bundle, a learning engine, and a detecting engine. The examination bundle includes at least one medical image of the subject from a first modality and at least one medical image of the subject from a second modality. The learning engine is employed for determining the characteristics of abnormalities within the at least one medical image from the first modality and the at least one medical image from the second modality. The detecting engine is employed for detecting abnormalities within at least one of the medical images comprising the examination bundle.
1. A system for the detection of abnormalities in a medical image of a subject, comprising:
an examination bundle including at least one medical image of the subject from a first modality and at least one medical image of the subject from at least a second modality;
a mapping engine for defining a correspondence relationship between the at least one medical image from the first modality and the at least one medical image from the at least second modality;
a training/learning engine for determining the characteristics of abnormalities using features extracted within the at least one medical image from the first modality and the at least one medical image from the at least second modality;
a classification engine for detecting abnormalities within at least one of the medical images comprising the examination bundle; and
a combining module combining training/learning processes for determining the characteristics of abnormalities using features extracted from the first and second modalities and the detected abnormalities from the first and second modalities using the classification engine for diagnosis and identifying abnormalities detected in both of the first and second modalities.
2. The system of claim 1 , wherein the training/learning engine registers the at least one medical image from the first modality with the at least one medical image from the second modality.
3. The system of claim 2 , wherein the training/learning engine performs parametric image registration, non-parametric image registration, intensity-based registration, correlation-based registration, or mutual information-based registration.
4. The system of claim 1 , wherein the training/learning engine registers at least one medical image from at least one modality to an atlas image.
5. The system of claim 1 , wherein the training/learning engine comprises (a) a first learning module for determining the characteristics of abnormalities within the at least one medical image from the first modality and (b) a second learning module for determining the characteristics of abnormalities within the at least one medical image from the second modality; and
wherein the classification engine comprises (a) a first detecting module for detecting abnormalities within the at least one medical image from the first modality, and (b) a second detecting module for detecting abnormalities within the at least one medical image from the second modality.
6. The system of claim 5 , wherein at least one of the first or second learning modules incorporates predetermined knowledge of the characteristics of abnormalities.
7. The system of claim 5 , wherein the classification engine further comprises a combining module for combining the detected abnormalities from the first detecting module with the detected abnormalities from the second detecting module.
8. A system for the detection of abnormalities in a medical image of a subject, comprising:
an examination bundle including at least one medical image of the subject from a first modality and at least one medical image of the subject from at least a second modality;
a plurality of trained images employed for determining the characteristics of abnormalities within the at least one medical image from the first modality and the at least one medical image from the at least second modality;
a classification engine for detecting abnormalities within the medical images of the first and second modalities comprising the examination bundle using the characteristics; and
a combining module combining the detected abnormalities for determining the characteristics of abnormalities using features extracted from the first and second modalities and the detected abnormalities of the classification engine for diagnosis and identifying abnormalities detected in both of the first and second modalities.
9. The system of claim 8 , further comprising a mapping engine for defining a correspondence relationship between the at least one medical image from the first modality and the at least one medical image from the second modality.
10. The system of claim 1 , wherein the combining module determines joint characteristics from both the first and second modalities.
11. The system of claim 1 , wherein the combining module identifies all detected abnormalities, identifies abnormalities detected by one modality, identifies abnormalities detected by both modalities and identifies abnormalities detected by a chosen modality.
12. A method, comprising:
determining characteristics of abnormalities in medical images of plural modalities using a training/learning engine;
detecting and classifying abnormalities in the images using a classification engine; and
combining the classified abnormalities and image features of the medical images of the plural modalities for use in computer aided analysis and diagnosis applied to an examination image bundle including patient images in the first and second modalities and one of identifying an abnormality detected in more than one of the plural modalities, and identifying an abnormality detected in a chosen modality and another one of the plural modalities.