Prospective abnormal shadow detecting system and method of and apparatus for judging whether prospective abnormal shadow is malignant or benignant

1. A prospective abnormal shadow detecting system comprising a prospective abnormal shadow detecting means which detects a prospective abnormal shadow in an image on the basis of image data representing the image and a prospective abnormal shadow information output means which outputs information on the prospective abnormal shadow detected by the prospective abnormal shadow detecting means, wherein the improvement comprises that

a malignancy judging means which judges whether the prospective abnormal shadow detected by the prospective abnormal shadow detecting means is malignant or benignant is provided, and

the prospective abnormal shadow information output means outputs the information on the prospective abnormal shadow in such a manner that whether the prospective abnormal shadow is malignant or benignant as judged by the malignancy judging means can be distinguished, and

in which the prospective abnormal shadow information output means outputs an index of the malignancy and/or an index of the benignancy of the prospective abnormal shadow, wherein the index represents a margin by which a feature value exceeds a threshold value.

2. A prospective abnormal shadow detecting system as defined in claim 1 in which the malignancy judging means judges whether the prospective abnormal shadow detected by the prospective abnormal shadow detecting means is malignant or benignant by

obtaining a benignancy evaluation function value which is a value of an evaluation function for benignancy, a malignancy evaluation function value which is a value of an evaluation function for malignancy, and a normalcy evaluation function value which is a value of an evaluation function for normalcy by defining feature values of the image data by a predetermined function, and

comparing the normalcy evaluation function value with the benignancy evaluation function value and the normalcy evaluation function value with the malignancy evaluation function value.

3. A prospective abnormal shadow detecting system as defined in claim 1 , wherein whether the prospective abnormal shadow is malignant or benignant as judged by the malignancy judging means is distinguished visually.

4. A prospective abnormal shadow detecting system as defined in claim 1 , wherein the information on the prospective abnormal shadow output by the prospective abnormal shadow information output means comprises a visible marking.

5. A prospective abnormal shadow detecting system comprising a prospective abnormal shadow detecting means which detects a prospective abnormal shadow in an image on the basis of image data representing the image and a prospective abnormal shadow information output means which outputs information on the prospective abnormal shadow detected by the prospective abnormal shadow detecting means, wherein the improvement comprises that

a malignancy judging means which judges whether the prospective abnormal shadow detected by the prospective abnormal shadow detecting means is malignant or benignant is provided, and

the prospective abnormal shadow information output means outputs the information on the prospective abnormal shadow in such a manner that whether the prospective abnormal shadow is malignant or benignant as judged by the malignancy judging means can be distinguished, and

in which the malignancy judging means judges whether the prospective abnormal shadow detected by the prospective abnormal shadow detecting means is malignant or benignant by

obtaining a benignancy evaluation function value which is a value of an evaluation function for benignancy, a malignancy evaluation function value which is a value of an evaluation function for malignancy, and a normalcy evaluation function value which is a value of an evaluation function for normalcy by defining feature values of the image data by a predetermined function, and

comparing the normalcy evaluation function value with the benignancy evaluation function value and the normalcy evaluation function value with the malignancy evaluation function value, and

in which the normalcy evaluation function value is the Mahalanobis distance from a normal tissue distribution, the benignancy evaluation function value is the Mahalanobis distance from a benignant tissue distribution, and the malignancy evaluation function value is the Mahalanobis distance from a malignant tissue distribution.

6. A prospective abnormal shadow detecting system as defined in claim 5 , wherein the prospective abnormal shadow information output means outputs an index of the malignancy and/or an index of the benignancy of the prospective abnormal shadow.

7. A method of judging whether a prospective abnormal shadow detected on the basis of image data representing an image is malignant or benignant comprising the steps of

obtaining a benignancy evaluation function value which is a value of an evaluation function for benignancy, a malignancy evaluation function value which is a value of an evaluation function for malignancy, and a normalcy evaluation function value which is a value of an evaluation function for normalcy by defining feature values of the image data by a predetermined function, and

comparing the normalcy evaluation function value with the benignancy evaluation function value and the normalcy evaluation function value with the malignancy evaluation function values,

wherein the normalcy evaluation function value is the Mahalanobis distance from a normal tissue distribution, the benignancy evaluation function value is the Mahalanobis distance from a benignant tissue distribution, and the malignancy evaluation function value is the Mahalanobis distance from a malignant tissue distribution.

8. An apparatus for judging whether a prospective abnormal shadow detected on the basis of image data representing an image is malignant or benignant comprising

an evaluation function value calculating means which obtains a benignancy evaluation function value which is a value of an evaluation function for benignancy, a malignancy evaluation function value which is a value of an evaluation function for malignancy, and a normalcy evaluation function value which is a value of an evaluation function for normalcy by defining feature values of the image data by a predetermined function, and

judging means which judges whether the prospective abnormal shadow is malignant or benignant by comparing the normalcy evaluation function value with the benignancy evaluation function value and the normalcy evaluation function value with the malignancy evaluation function values,

wherein the evaluation function calculating means calculates the Mahalanobis distance from a normal tissue distribution as the normalcy evaluation function value, the Mahalanobis distance from a benignant tissue distribution as the benignancv evaluation function value, and the Mahalanobis distance from a malignancy tissue distribution as the malignancy evaluation function value.

9. A prospective abnormal shadow detecting system comprising a prospective abnormal shadow detecting means which detects a prospective abnormal shadow in an image on the basis of image data representing the image and a prospective abnormal shadow information output means which outputs information on the prospective abnormal shadow detected by the prospective abnormal shadow detecting means, wherein the improvement comprises that

a malignancy judging means which judges whether the prospective abnormal shadow detected by the prospective abnormal shadow detecting means is malignant or benignant is provided, and

the prospective abnormal shadow information output means outputs the information on the prospective abnormal shadow in such a manner that whether the prospective abnormal shadow is malignant or benignant as judged by the malignancy judging means can be distinguished, and

in which the malignancy judging means judges whether the prospective abnormal shadow detected by the prospective abnormal shadow detecting means is malignant or benignant by

obtaining a benignancy evaluation function value which is a value of an evaluation function for benignancy, a malignancy evaluation function value which is a value of an evaluation function for malignancy, and a normalcy evaluation function value which is a value of an evaluation function for normalcy by defining feature values of the image data by a predetermined function, and

comparing the normalcy evaluation function value with the benignancy evaluation function value and the normalcy evaluation function value with the malignancy evaluation function value, and

wherein said malignancy judging means calculates margins of said feature values with respect to respective threshold values of said feature values, and outputs said feature values and said margins.

10. A prospective abnormal shadow detecting system as defined in claim 9 , wherein the prospective abnormal shadow information output means outputs an index of the malignancy and/or an index of the benignancy of the prospective abnormal shadow.

11. A prospective abnormal shadow detecting system comprising a prospective abnormal shadow detecting means which detects a prospective abnormal shadow in an image on the basis of image data representing the image and a prospective abnormal shadow information output means which outputs information on the prospective abnormal shadow detected by the prospective abnormal shadow detecting means, wherein the improvement comprises that

a malignancy judging means which judges whether the prospective abnormal shadow detected by the prospective abnormal shadow detecting means is malignant or benignant is provided, and

the prospective abnormal shadow information output means outputs the information on the prospective abnormal shadow in such a manner that whether the prospective abnormal shadow is malignant or benignant as judged by the malignancy judging means can be distinguished, and

in which the malignancy judging means judges whether the prospective abnormal shadow detected by the prospective abnormal shadow detecting means is malignant or benignant by

obtaining a benignancy evaluation function value which is a value of an evaluation function for benignancy, a malignancy evaluation function value which is a value of an evaluation function for malignancy, and a normalcy evaluation function value which is a value of an evaluation function for normalcy by defining feature values of the image data by a predetermined function, and

comparing the normalcy evaluation function value with the benignancy evaluation function value and the normalcy evaluation function value with the malignancy evaluation function value, and

wherein said prospective abnormal shadow information output means displays an overall image, a processed interest area image, said feature values and calculated margins of said feature values in one frame.

12. A prospective abnormal shadow detecting system as defined in claim 11 , wherein the prospective abnormal shadow information output means outputs an index of the malignancy and/or an index of the benignancy of the prospective abnormal shadow.