Treatment adherence systems and processes

1. A method, comprising:

receiving, by a computer device, an association of a prescribed treatment of an ailment to a user;

receiving, by the computer device, an image of the user;

receiving, by the computer device, an image of treatment adherence by the user;

determining, by the computer device, adherence to the prescribed treatment by analyzing the image of treatment adherence, wherein the determining comprises extracting features from the image of treatment adherence using photographic interpretation techniques and comparing the extracted features to known features of the prescribed treatment; and

generating, by the computer device, a personalized visualization illustrating the determined adherence to the prescribed treatment by augmenting the image of the user in the personalized visualization with images of the ailment of the user based on the determined adherence,

wherein the generating the personalized visualization is performed using a deep generative machine learning model trained with images of treatment adherence, the deep generative machine learning model basing the personal visualization on the image of treatment adherence that is previously unseen data, and the deep generative machine learning model utilizing variational autoencoders that reconstruct datasets using autoencoding and decoding in view of information within the datasets.

2. The method of claim 1 , further comprising sending, by the computer device, a message containing the personalized visualization to the user via a user device.

3. The method of claim 1 , further comprising storing, by the computer device, adherence data defining the determined adherence to the prescribed treatment in a database.

4. The method of claim 1 , further comprising generating, by the computer device, an initial visualization showing results of adherence or nonadherence to the prescribed treatment prior to a capturing of the image of treatment adherence.

5. The method of claim 1 , wherein the analyzing the image of the treatment adherence includes analyzing a geometry and a symmetry of a pill within the image.

6. The method of claim 1 , further comprising comparing, by the computer device, the extracted features to known features of the prescribed treatment by extracting number features from the image that indicate an amount of medicine of the prescribed treatment.

7. The method of claim 1 , further comprising modifying, by the computer device, the personalized visualization in view of nonadherence to the prescribed treatment by augmenting the personalized visualization with images of a disease of the user at various stages, the disease being associated with the prescribed treatment.

8. The method of claim 2 , further comprising monitoring, by the computer device, further adherence or nonadherence to the prescribed treatment after the sending the message containing the personalized visualization to the user.

9. The method of claim 8 , further comprising training, by the computer device, the deep generative machine learning model by assigning coordinates to features within images of ailments.

10. The method of claim 9 , further comprising providing, by the computer device, a nonadherence notification to a provider of the prescribed treatment in response to the nonadherence to the prescribed treatment.

11. The method of claim 10 , wherein the personalized visualization includes computer enhanced items showing results of the determined adherence to the prescribed treatment.

12. The method of claim 11 , wherein the deep generative machine learning model includes generative adversarial networks.

13. A computer program product comprising one or more computer readable storage media having program instructions collectively stored on the one or more computer readable storage media, the program instructions executable to:

train deep generative machine learning models to learn a physical progression of ailments and healing of the ailments for a prescribed treatment;

receive an association of the prescribed treatment with a user;

receive an image of the user;

generate an initial visualization by modifying the image of the user using the deep generative machine learning models;

send the initial visualization to a computing device of the user for display on the computing device of the user;

receive an image of treatment adherence by the user;

determine adherence to the prescribed treatment by extracting features from the image of treatment adherence and comparing the extracted features to known features of the prescribed treatment; and

generate a modified personalized visualization based on the determined adherence and augmenting the image of the user with a hypothetical image or illustration based on the deep generative machine learning models,

wherein the deep generative machine learning models are trained with images of treatment adherence, the deep generative machine learning models base the modified personalized visualization on the image of treatment adherence that is previously unseen data, and the deep generative machine learning models utilize variational autoencoders that reconstruct datasets using autoencoding and decoding in view of information within the datasets.

14. The computer program product of claim 13 , wherein the deep generative machine learning models include utilizing the variational autoencoders on images of the human body in different states of health or ailment based on the physical progression of ailments.

15. The computer program product of claim 13 , wherein the modified personalized visualization illustrates computer enhanced items showing consequences for the nonadherence to the prescribed treatment.

16. The computer program product of claim 13 , wherein the program instructions are executable to send a message to the computing device of the user, the message containing the modified personalized visualization.

17. A system comprising:

a processor, a computer readable memory, one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable to:

train deep generative machine learning models to learn a physical progression of ailments and healing of the ailments for a prescribed treatment;

receive an association of the prescribed treatment with a user;

receive an image of the user commencing the prescribed treatment;

generate an initial visualization by modifying the image of the user using the deep generative machine learning models to illustrate results of nonadherence to the prescribed treatment including enhanced changes to the image of the user showing consequences for nonadherence on a virtual pictorial and/or animated representation illustrating the ailment;

send a message to the computing device of the user which contains the initial visualization;

receive an image of treatment adherence by the user;

determine adherence to the prescribed treatment by extracting features from the image of treatment adherence and comparing the extracted features to known features of the prescribed treatment; and

generate a personalized visualization in response to adherence to the prescribed treatment,

wherein the deep generative machine learning models are trained with images of treatment adherence, the deep generative machine learning models base the personalized visualization on the image of treatment adherence that is previously unseen data, and the deep generative machine learning models utilize variational autoencoders that reconstruct datasets using autoencoding and decoding in view of information within the datasets.

18. The system of claim 17 , wherein the personalized visualization illustrates computer enhanced items showing results of the adherence to the prescribed treatment.

19. The system of claim 17 , wherein the deep generative machine learning models include utilizing the variational autoencoders on images of the human body in different states of health or ailment based on the physical progression of ailments.

20. The system of claim 17 , wherein the deep generative machine learning models include generative adversarial networks on images of the human body in different states of health or ailment.