Methods and apparatus for the application of machine learning to radiographic images of animals

1. A computer-implemented method of training a plurality of segmentation artificial intelligence engines for segmentation of an anatomical area of an animal, comprising:

generating a plurality of segmentation training sets, the plurality of segmentation training sets comprising a first segmentation training set and a second segmentation training set;

training a first segmentation artificial intelligence engine of the plurality of segmentation artificial intelligence engines using the first segmentation training set;

training a second segmentation artificial intelligence engine of the plurality of segmentation artificial engines using the first segmentation training set and the second segmentation training set;

receiving a set of radiographic images captured of the animal;

segmenting the set of radiographic images captured of the animal into the anatomical area of the animal using the first segmentation artificial intelligence engine and the second segmentation artificial intelligence engine to create a segmented set of radiographic images, at least a portion of the segmented set of radiographic images segmented by the first segmentation artificial intelligence engine comprising a plurality of distinct anatomical areas for the animal as compared with the segmented set of radiographic images segmented by the second segmentation artificial intelligence engine; and

feeding the created segmented set of radiographic images from the first segmentation artificial intelligence engine and the second segmentation artificial intelligence engine to one or more classification artificial intelligence engines.

2. The computer-implemented method of claim 1 , further comprising characterizing each of the plurality of segmentation training sets dependent upon animal species.

3. The computer-implemented method of claim 2 , further comprising characterizing each of the plurality of segmentation training sets dependent upon animal breed.

4. The computer-implemented method of claim 3 , further comprising determining the animal species and the animal breed of the animal by reading metadata associated with the set of radiographic images captured of the animal; and

using the determining of the animal species and the animal breed of the animal in order to select one or more of the plurality of segmentation training sets for the segmenting of the set of radiographic images captured of the animal.

5. The computer-implemented method of claim 4 , further comprising feeding the segmented set of radiographic images of the animal into one or more classification artificial intelligence engines in order to classify a plurality of conditions for the animal.

6. The computer-implemented method of claim 5 , further comprising determining a confidence level for the plurality of conditions for the animal that have been classified.

7. The computer-implemented method of claim 1 , further comprising adding the segmented set of radiographic images to the first segmentation training set and/or the second segmentation training set.

8. A non-transitory computer-readable storage apparatus comprising a plurality of instructions, that when executed by a processor apparatus, are configured to:

generate a plurality of segmentation training sets, the plurality of segmentation training sets comprising a first segmentation training set and a second segmentation training set;

train a first segmentation artificial intelligence engine of a plurality of segmentation artificial intelligence engines using the first segmentation training set;

train a second segmentation artificial intelligence engine of the plurality of segmentation artificial engines using the first segmentation training set and the second segmentation training set;

receive a set of radiographic images captured of an animal;

segment the set of radiographic images captured of the animal into the anatomical area of the animal using the first segmentation artificial intelligence engine and the second segmentation artificial intelligence engine to create a segmented set of radiographic images, at least a portion of the segmented set of radiographic images segmented by the first segmentation artificial intelligence engine comprising a plurality of distinct anatomical areas for the animal as compared with the segmented set of radiographic images segmented by the second segmentation artificial intelligence engine; and

feed the created segmented set of radiographic images from the first segmentation artificial intelligence engine and the second segmentation artificial intelligence engine to one or more classification artificial intelligence engines.

9. The non-transitory computer-readable storage apparatus of claim 8 , wherein the plurality of instructions, when executed by the processor apparatus, are further configured to:

characterize each of the plurality of segmentation training sets dependent upon animal species.

10. The non-transitory computer-readable storage apparatus of claim 9 , wherein the plurality of instructions, when executed by the processor apparatus, are further configured to:

characterize each of the plurality of segmentation training sets dependent upon animal breed.

11. The non-transitory computer-readable storage apparatus of claim 10 , wherein the plurality of instructions, when executed by the processor apparatus, are further configured to:

read metadata associated with the set of radiographic images captured of the animal in order to determine the animal species and the animal breed of the animal; and

use the determination of the animal species and the animal breed of the animal in order to select one or more of the plurality of segmentation training sets for the segmentation of the set of radiographic images captured of the animal.

12. The non-transitory computer-readable storage apparatus of claim 11 , wherein the plurality of instructions, when executed by the processor apparatus, are further configured to:

feed the segmented set of radiographic images of the animal into one or more classification artificial intelligence engines in order to classify a plurality of conditions for the animal.

13. The non-transitory computer-readable storage apparatus of claim 12 , wherein the plurality of instructions, when executed by the processor apparatus, are further configured to:

determine a confidence level for the plurality of conditions for the animal that have been classified.

14. The non-transitory computer-readable storage apparatus of claim 8 , wherein the plurality of instructions, when executed by the processor apparatus, are further configured to:

add the segmented set of radiographic images to the first segmentation training set and/or the second segmentation training set.

15. A system for segmenting radiographic images of an animal, the system comprising:

a plurality of segmentation artificial intelligence engines and a data storage apparatus, wherein the system is configured to:

generate a plurality of segmentation training sets and store the plurality of segmentation training sets in the data storage apparatus, the plurality of segmentation training sets comprising a first segmentation training set and a second segmentation training set;

train a first segmentation artificial intelligence engine of the plurality of segmentation artificial intelligence engines using the first segmentation training set;

train a second segmentation artificial intelligence engine of the plurality of segmentation artificial engines using the first segmentation training set and the second segmentation training set;

receive a set of radiographic images captured of an animal from the data storage apparatus;

segment the set of radiographic images captured of the animal into the anatomical area of the animal using one or both of the first segmentation artificial intelligence engine and the second segmentation artificial intelligence engine to create a segmented set of radiographic images, at least a portion of the segmented set of radiographic images segmented by the first segmentation artificial intelligence engine comprising a plurality of distinct anatomical areas for the animal as compared with the segmented set of radiographic images segmented by the second segmentation artificial intelligence engine; and

feed the created segmented set of radiographic images from the first segmentation artificial intelligence engine and the second segmentation artificial intelligence engine to one or more classification artificial intelligence engines.

16. The system of claim 15 , wherein the system is further configured to:

characterize each of the plurality of segmentation training sets dependent upon animal species.

17. The system of claim 16 , wherein the system is further configured to:

characterize each of the plurality of segmentation training sets dependent upon animal breed.

18. The system of claim 17 , wherein the system is further configured to:

read metadata associated with the set of radiographic images captured of the animal in order to determine the animal species and the animal breed of the animal; and

use the determination of the animal species and the animal breed of the animal in order to select one or more of the plurality of segmentation training sets for the segmentation of the set of radiographic images captured of the animal.

19. The system of claim 18 , wherein the system is further configured to:

feed the segmented set of radiographic images of the animal into one or more classification artificial intelligence engines in order to classify a plurality of conditions for the animal.

20. The system of claim 15 , wherein the system is further configured to:

add the segmented set of radiographic images to the first segmentation training set and/or the second segmentation training set.