Estimating spinal age

1. A computer-implemented method, the computer-implemented method comprising:

receiving, by one or more computer processors, image data of a subject including at least a portion of a spine of the subject and a chronological age of the subject;

pre-processing, by one or more computer processors, the image data including at least the portion of a spine; and

determining, by one or more computer processors, an apparent age of one of the spine or the portion of the spine of the subject using a trained artificial intelligence deep learning algorithm at least by:

retrieving, by one or more computer processors, a plurality of spine image data of a plurality of subjects for training the artificial intelligence deep learning algorithm;

pre-processing, by one or more computer processors, the plurality of the spine image data of the plurality of subjects; and

training, by one or more computer processors, a machine learning algorithm using unsupervised deep learning wherein, the machine learning algorithm is a deep learning network wherein, a latent variable indicates an apparent spine age, the chronological age of the subject is a random variable derived from a distribution of a plurality of probabilities of the chronological age with respect to the apparent age and wherein the deep learning network learns the distribution of the plurality of probabilities chronological age with respect to the apparent age from the training using the plurality of image data.

2. The computer-implemented method of claim 1 , wherein receiving the image data of the subject including at least a portion of the spine of the subject includes receiving one of a computed tomography (CT) scan, a CT volume, or a magnetic resonance imaging (MRI) scan of at least a portion of the spine of the subject.

3. The computer-implemented method of claim 2 , wherein pre-processing the image data including at least a portion of the spine of the subject includes extracting one of the spine or the portion of the spine from one of the CT scan, the CT volume, or a MRI scan.

4. The computer-implemented method of claim 3 , wherein the spine or the portion of the spine extracted from the image data is one of a two-dimensional section or a three-dimensional section of the spine or the portion of the spine, and wherein the portion of the spine includes one of a vertebra, a disc, or a combination of more than one vertebra and one or more disc of the spine of the subject.

5. The computer-implemented method of claim 1 , wherein training the machine learning algorithm using unsupervised deep learning includes using a deep learning network with an additional softmax layer on top of a last densely connected softmax layer.

6. The computer-implemented method of claim 5 , wherein training the machine learning algorithm using unsupervised deep learning comprises adding an inverse deep learning network as an autoencoder.

7. The computer-implemented method of claim 1 , wherein training the machine learning algorithm using unsupervised deep learning includes using soft labelling and minimum entropy with a softmax regression.

8. The computer-implemented method of claim 1 , further comprising adding a notation of excessive spinal degeneration associated with one of the spine or the portion of the spine to an output in response to the apparent age of one of the spine or a portion of the spine of the subject is greater than the chronological age of the subject by more than a pre-set number of years.

9. A computer program product for determining an apparent spine age, the computer program product comprising:

one or more computer readable storage media; and

program instructions stored on the one or more computer readable storage media, the program instructions executable by a processor, the program instructions comprising instructions for:

receiving image data of a subject including at least a portion of a spine of the subject and a chronological age of the subject;

pre-processing the image data including at least the portion of a spine; and

determining an apparent age of one of the spine or the portion of the spine of the subject using a trained artificial intelligence deep learning algorithm at least by:

retrieving, by one or more computer processors, a plurality of spine image data of a plurality of subjects for training the artificial intelligence deep learning algorithm;

pre-processing, by one or more computer processors, the plurality of the spine image data of the plurality of subjects; and

training, by one or more computer processors, a machine learning algorithm using unsupervised deep learning wherein, the machine learning algorithm is a deep learning network wherein, a latent variable indicates an apparent spine age, the chronological age of the subject is a random variable derived from a distribution of a plurality of probabilities of the chronological age with respect to the apparent age and wherein the deep learning network learns the distribution of the plurality of probabilities chronological age with respect to the apparent age from the training using the plurality of image data.

10. The computer program product of claim 9 , wherein receiving the image data of the subject including at least a portion of the spine of the subject includes receiving one of a computed tomography (CT) scan, a CT volume, or a magnetic resonance imaging (MRI) scan of at least a portion of the spine of the subject.

11. The computer program product of claim 10 , wherein pre-processing the image data including at least a portion of the spine of the subject includes extracting one of the spine or the portion of the spine from one of the CT scan, the CT volume, or a MRI scan.

12. The computer program product of claim 11 , wherein the spine or the portion of the spine extracted from the image data is one of a two-dimensional section or a three-dimensional section of the spine or the portion of the spine, and wherein the portion of the spine includes one of a vertebra, a disc, or a combination of more than one vertebra and one or more disc of a spine of the subject.

13. The computer program product of claim 9 , wherein training the machine learning algorithm using unsupervised deep learning includes using a deep learning network with an additional softmax layer on top of a last densely connected softmax layer.

14. The computer program product of claim 13 , wherein training the machine learning algorithm using unsupervised deep learning comprises adding an inverse deep learning network as an autoencoder.

15. The computer program product of claim 9 , wherein training the machine learning algorithm using unsupervised deep learning includes using soft labelling and minimum entropy with a softmax regression.

16. The computer program product of claim 9 , further comprising adding a notation of excessive spinal degeneration associated with one of the spine or the portion of the spine to an output in response to the apparent age of one of the spine or a portion of the spine of the subject is greater than the chronological age of the subject by more than a pre-set number of years.

17. A computer system comprising:

one or more computer processors;

one or more computer readable storage media;

program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more processors, the program instructions comprising instructions to perform:

receiving image data of a subject including at least a portion of a spine of the subject and a chronological age of the subject;

pre-processing the image data including at least the portion of a spine; and

determining an apparent age of one of the spine or the portion of the spine of the subject using a trained artificial intelligence deep learning algorithm at least by:

retrieving, by one or more computer processors, a plurality of spine image data of a plurality of subjects for training the artificial intelligence deep learning algorithm;

pre-processing, by one or more computer processors, the plurality of the spine image data of the plurality of subjects; and

training, by one or more computer processors, a machine learning algorithm using unsupervised deep learning wherein, the machine learning algorithm is a deep learning network wherein, a latent variable indicates an apparent spine age, the chronological age of the subject is a random variable derived from a distribution of a plurality of probabilities of the chronological age with respect to the apparent age and wherein the deep learning network learns the distribution of the plurality of probabilities chronological age with respect to the apparent age from the training using the plurality of image data.

18. The computer system of claim 17 , wherein training the machine learning algorithm using unsupervised deep learning includes using a deep learning network with an additional softmax layer on top of a last densely connected softmax layer.

19. The computer system of claim 17 , wherein training the machine learning algorithm using unsupervised deep learning includes using soft labelling and minimum entropy with a softmax regression.

20. The computer system of claim 17 , further comprising adding a notation of excessive spinal degeneration associated with one of the spine or the portion of the spine to an output in response to the apparent age of one of the spine or a portion of the spine of the subject is greater than the chronological age of the subject by more than a pre-set number of years.