Digital medical image analysis
A medical image is analyzed using an algorithm requiring input parameters. Values of the input parameters are derived from metadata, indicating properties of the medical image. For example, the metadata may indicate the type of image acquisition device or settings used to create the medical image. In another example, the metadata may relate to the patient upon whom the image is based. This allows optimum values to be selected for the parameters.
1. A method of analyzing a digital medical image using a computer-implemented algorithm having one or more variable input parameters, comprising:
a. receiving the digital medical image and metadata associated with the digital medical image;
b. setting one or more parameter values corresponding to one or more of the variable input parameters based on the metadata; and
c. processing the digital medical image using the algorithm and the one or more parameter values set in step b.
2. The method of claim 1 , wherein receiving the digital medical image and the metadata includes receiving metadata that is indicative of an attribute of a patient upon whom the image is based.
3. The method of claim 2 , wherein receiving the digital medical image and the metadata includes receiving metadata that is indicative of the attribute that includes at least one of a clinical indication, historical clinical information, and statistical information.
4. The method of claim 1 , wherein receiving the digital medical image and the metadata includes receiving metadata that is indicative of an attribute of an imaging device used to obtain the image.
5. The method of claim 4 , wherein receiving the digital medical image and the metadata includes receiving metadata that is indicative of the attribute that at least one of a type of the imaging device and a setting of the imaging device.
6. The method of claim 1 , wherein setting the one or more parameter values includes selecting one or more parameter values from a plurality of parameter value sets.
7. The method of claim 1 , wherein setting the one or more parameter values includes deriving the one or more parameter values as a predetermined function of the metadata.
8. The method of claim 1 , wherein setting the one or more parameter values includes deriving the one or more parameter values from the metadata using at least one of a linear filter, a nonlinear filter, fuzzy logic, and a neural network.
9. The method of claim 1 , wherein setting the one or more parameter values includes modifying the one or more parameter values in response to receiving user input to provide modified one or more parameter values, and wherein processing the digital medical image is performed using the algorithm and the modified one or more parameter values.
10. The method of claim 9 , further comprising:
displaying a result of the step of processing the digital medical image and the modified one or more parameter values.
11. The method of claim 10 , wherein displaying the result includes displaying the result that includes a modified version of the digital medical image to enable a visual comparison of the digital medical image and the modified version.
12. The method of claim 11 , wherein displaying the result includes displaying the digital medical image and the modified image simultaneously.
13. The method of claim 11 , wherein displaying the result includes displaying the digital medical image and the modified image alternately.
14. A method of deriving a set of one or more optimum parameter values to be provided to a computer-implemented algorithm to analyze a candidate digital medical image having associated metadata, comprising:
a. accessing a plurality of training digital medical image data sets each including metadata associated with a corresponding digital medical image;
b. deriving a relationship between values of the metadata and optimum parameter values of corresponding training digital medical image data sets; and
c. deriving, based on the relationship and on metadata of the candidate digital medical image, the set of optimum parameter values for the candidate digital medical image.
15. The method of claim 14 , wherein accessing the plurality of training digital medical image data sets includes accessing training digital medical image data sets, each set including metadata that is indicative of an attribute of a patient upon whom the image is based.
16. The method of claim 14 , wherein accessing the plurality of training digital medical image data sets includes accessing training digital medical image data sets, each set including metadata that is indicative of an attribute of an imaging device used to obtain the image.
17. The method of claim 14 , further comprising:
inputting the metadata by a user.
18. The method of claim 14 , wherein deriving the relationship includes deriving a relationship corresponding to a selection from a plurality of parameter value sets based on the values of the metadata.
19. The method of claim 14 , wherein deriving the relationship includes deriving a relationship that includes a predetermined function based on the metadata.
20. The method of claim 14 , wherein deriving the relationship includes deriving a relationship that is based on a neural network.
21. The method of claim 14 , wherein deriving the relationship includes deriving a relationship that includes a fuzzy logic relationship.
22. The method of claim 14 , wherein accessing the plurality of training digital medical image data sets includes locally accessing training digital medical image data sets.
23. The method of claim 14 , wherein accessing the plurality of training digital medical image data sets includes accessing training digital medical image data sets over a network.
24. The method of claim 23 , wherein accessing the plurality of training digital medical image data sets includes accessing training digital medical image data sets over a network that includes at least one of a local area network, a wide area network, and a PACS network.
25. The method of claim 14 , wherein accessing the plurality of training digital medical image data sets includes selecting training digital medical image data sets for access using a predetermined criterion.
26. The method of claim 14 , wherein deriving the relationship includes weighting the training image data sets.
27. The method of claim 14 , wherein deriving the relationship includes receiving as input a user preference for optimization of the parameter values.
28. An article comprising a computer readable medium for storing instructions to enable a processor-based system to perform operations to analyze a digital medical image using a computer-implemented algorithm having one or more variable input parameters, the operations comprising:
receiving the digital medical image and metadata associated with the digital medical image;
setting one or more parameter values of corresponding one or more of the variable input parameters based on the metadata; and
processing the digital medical image using the algorithm and the one or more parameter values.
29. An article comprising a computer readable medium for storing instructions to enable a processor-based system to perform operations to derive a set of one or more optimum parameter values to be provided to a computer-implemented algorithm to analyze a candidate digital medical image having associated metadata, the operations comprising:
accessing a plurality of training digital medical image data sets each including metadata associated with a corresponding digital medical image;
deriving a relationship between values of the metadata and optimum parameter values of corresponding training digital medical image data sets; and
deriving, based on the relationship and on metadata of the candidate digital medical image, the set of optimum parameter values for the candidate digital medical image.