Method of detecting abnormal pattern candidates

1. A method of detecting an abnormal pattern candidate, in which processing for detecting an abnormal pattern candidate embedded in a medical image having been obtained from an image recording operation is performed on an image signal representing the medical image, the method comprising the steps of:

i) in cases where the image signal is an image signal which has been subjected to predetermined signal processing, performing correction processing on the image signal such that the abnormal pattern candidate detecting processing performed on the image signal, which has been subjected to the predetermined signal processing, does not depend upon the predetermined signal processing, the correction processing is performed such that a level of detection of an abnormal pattern candidate is kept at a predetermined level, and

ii) performing the abnormal pattern candidate detecting processing on the image signal, which has been obtained from the correction processing.

2. A method as defined in claim 1 wherein the predetermined signal processing is signal processing performed in accordance with an index value, which represents an imaging environment of the medical image.

3. A method as defined in claim 2 wherein a threshold value, which is utilized in the abnormal pattern candidate detecting processing in making a judgment as to whether a pattern is or is not the abnormal pattern candidate, is corrected in accordance with the index value, which represents the imaging environment of the medical image.

4. A method as defined in claim 1 , 2 , or 3 wherein the abnormal pattern candidate detecting processing is a processing, in which iris filter processing is performed on the image signal, and a tumor pattern candidate is thereby detected as the abnormal pattern candidate.

5. A method as defined in claim 1 , 2 , or 3 wherein the abnormal pattern candidate detecting processing is a processing, in which morphological operation processing is performed on the image signal, and a microcalcification pattern candidate is thereby detected as the abnormal pattern candidate.

6. The method of detecting an abnormal pattern candidate as defined in claim 1 , wherein the predetermined image processing is a normalization process.

7. The method of detecting an abnormal pattern candidate as defined in claim 6 , wherein the normalization process is exposure data recognizer (EDR) processing.

8. The method of detecting an abnormal pattern candidate as defined in claim 7 , wherein a mean value of image signal values corresponding to a specific region of the image signal are used in the EDR processing.

9. A method of detecting an abnormal pattern candidate, in which processing for detecting an abnormal pattern candidate embedded in a medical image having been obtained from an image recording operation is performed on an image signal representing the medical image, the method comprising the steps of:

i) in cases where the image signal is an image signal which has been subjected to predetermined signal processing, performing correction processing on the image signal such that the abnormal pattern candidate detecting processing performed on the image signal, which has been subjected to the predetermined signal processing, does not depend upon the predetermined signal processing, and

ii) performing the abnormal pattern candidate detecting processing on the image signal, which has been obtained from the correction processing,

wherein the predetermined signal processing is signal processing performed in accordance with an index value, which represents an imaging environment of the medical image, and

wherein the signal processing is normalization processing, which may be represented by Formula (1), and the correction processing is a processing, which may be represented by Formula (2):

Q out=( RQ/Rq )( Q in− qm )/ L+Qm   (1)

Q out′=( L/L 0)( Q out− Qm )+ Qm   (2)

wherein

Qin represents the image signal before being subjected to the signal processing,

Qout represents the image signal after being subjected to the signal processing,

qm represents the predetermined signal value before being subjected to the signal processing,

Qm represents the value which qm takes after the signal processing has been performed,

L=(Δq/Rq)/(ΔQ/RQ),

Δq represents the width (latitude) of the image signal before being subjected to the signal processing,

ΔQ represents the width (predetermined value) of the image signal after being subjected to the signal processing,

Rq represents the signal range of Qin,

RQ represents the signal range of Qout,

L 0 represents the reference L value, and

Qout′ represents the image signal after being subjected to the correction processing.

10. A method as defined in claim 9 wherein a lower limit value of L is determined previously, and the value of L is set at the lower limit value of L in cases where the value of L in Formula (2) is at most equal to the lower limit value having been determined.

11. A method as defined in claim 10 , wherein an upper limit value of L is determined previously, and the value of L is set at the upper limit value of L in cases where the value of L in Formula (2) is at least equal to the upper limit value having been determined.

12. A method as defined in claim 9 , wherein an upper limit value of L is determined previously, and the value of L is set at the upper limit value of L in cases where the value of L in Formula (2) is at least equal to the upper limit value having been determined.

13. A method of detecting an abnormal pattern candidate, in which processing for detecting an abnormal pattern candidate embedded in a medical image having been obtained from an image recording operation is performed on an image signal representing the medical image, the method comprising the steps of:

i) in cases where the image signal is an image signal which has been subjected to predetermined signal processing, performing correction processing on the image signal such that the abnormal pattern candidate detecting processing performed on the image signal, which has been subjected to the predetermined signal processing, does not depend upon the predetermined signal processing, and

ii) performing the abnormal pattern candidate detecting processing on the image signal, which has been obtained from the correction processing,

wherein the predetermined signal processing is signal processing performed in accordance with an index value, which represents an imaging environment of the medical image,

wherein a threshold value, which is utilized in the abnormal pattern candidate detecting processing in making a judgment as to whether a pattern is or is not the abnormal pattern candidate, is corrected in accordance with the index value, which represents the imaging environment of the medical image, and

wherein the signal processing is normalization processing, which may be represented by Formula (1), and the correction processing is a processing, which may be represented by Formula (2):

Q out=( RQ/Rq )( Q in− qm )/ L+Qm   (1)

Q out′=( L/L 0)( Q out− Qm )+ Qm   (2)

wherein

Qin represents the image signal before being subjected to the signal processing,

Qout represents the image signal after being subjected to the signal processing,

qm represents the predetermined signal value before being subjected to the signal processing,

Qm represents the value which qm takes after the signal processing has been performed,

L=(Δq/Rq)/(ΔQ/RQ),

Δq represents the width (latitude) of the image signal before being subjected to the signal processing,

ΔQ represents the width (predetermined value) of the image signal after being subjected to the signal processing,

Rq represents the signal range of Qin,

RQ represents the signal range of Qout,

L 0 represents the reference L value, and

Qout′ represents the image signal after being subjected to the correction processing.

14. A method as defined in claim 13 wherein a lower limit value of L is determined previously, and the value of L is set at the lower limit value of L in cases where the value of L in Formula (2) is at most equal to the lower limit value having been determined.

15. A method as defined in claim 14 , wherein an upper limit value of L is determined previously, and the value of L is set at the upper limit value of L in cases where the value of L in Formula (2) is at least equal to the upper limit value having been determined.

16. A method as defined in claim 13 , wherein an upper limit value of L is determined previously, and the value of L is set at the upper limit value of L in cases where the value of L in Formula (2) is at least equal to the upper limit value having been determined.