Data aggregation, integration and analysis system and related devices and methods
A system for recording, storing and processing diagnostic information, including: a computer implementing a computer-readable media including digital data and ground truth; a registry constructed and arranged to store and associate transactions or accesses on the data; and a machine learning system that considers each learning step modification a microtransaction for the data used in that step and which is recorded in the transaction registry. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.
1. A method comprising:
applying an image associated as input to a machine learning model;
receiving, as output from the machine learning model, a prediction of a condition associated with the image; and
assigning attribution of the prediction to one or more of a plurality of training images, wherein:
the plurality of training images are registered on a distributed ledger, each training image labeled with truth data; and
the machine learning model is trained using the plurality of training images, wherein, as the machine learning model is trained, each of a plurality of modifications of the machine learning model that are attributable to a given training image are recorded as microtransactions to the distributed ledger in association with the given training image.
2. The method of claim 1 , wherein the machine learning model is a neural network.
3. The method of claim 1 , wherein the truth data for a given training image comprises diagnostic data associated with the given training image.
4. The method of claim 3 , wherein the prediction comprises a diagnosis of one or more medical conditions.
5. The method of claim 3 , wherein the given training image is an image of a patient, and wherein the diagnostic data comprises one or more medical conditions of the patient at a time when the image was captured.
6. The method of claim 5 , wherein the truth data comprises information about an entity that determined the diagnostic data based on interactions with the patient.
7. The method of claim 1 , wherein each respective modification of the plurality of modifications made to the machine learning model is registered on the distributed ledger, and wherein the respective modification to the training image that resulted in the modification is attributed to the training image that resulted in the respective modification.
8. A non-transitory computer-readable medium comprising memory with instructions encoded thereon, the instructions, when executed, causing one or more processors to perform operations, the instructions comprising instructions to:
applying an image associated as input to a machine learning model;
receiving, as output from the machine learning model, a prediction of a condition associated with the image; and
assigning attribution of the prediction to one or more of a plurality of training images, wherein:
the plurality of training images are registered on a distributed ledger, each training image labeled with truth data; and
the machine learning model is trained using the plurality of training images, wherein, as the machine learning model is trained, each of a plurality of modifications of the machine learning model that are attributable to a given training image are recorded as microtransactions to the distributed ledger in association with the given training image.
9. The non-transitory computer-readable medium of claim 8 , wherein the machine learning model is a neural network.
10. The non-transitory computer-readable medium of claim 8 , wherein the truth data for a given training image comprises diagnostic data associated with the given training image.
11. The non-transitory computer-readable medium of claim 10 , wherein the prediction comprises a diagnosis of one or more medical conditions.
12. The non-transitory computer-readable medium of claim 10 , wherein the given training image is an image of a patient, and wherein the diagnostic data comprises one or more medical conditions of the patient at a time when the image was captured.
13. The non-transitory computer-readable medium of claim 12 , wherein the truth data comprises information about an entity that determined the diagnostic data based on interactions with the patient.
14. The non-transitory computer-readable medium of claim 8 , wherein each respective modification of the plurality of modifications made to the machine learning model is registered on the distributed ledger, and wherein the respective modification to the training image that resulted in the modification is attributed to the training image that resulted in the respective modification.
15. A system comprising:
memory with instructions encoded thereon; and
one or more processors that, when executing the instructions, are caused to perform operations comprising:
applying an image associated as input to a machine learning model;
receiving, as output from the machine learning model, a prediction of a condition associated with the image; and
assigning attribution of the prediction to one or more of a plurality of training images, wherein:
the plurality of training images are registered on a distributed ledger, each training image labeled with truth data; and
the machine learning model is trained using the plurality of training images, wherein, as the machine learning model is trained, each of a plurality of modifications of the machine learning model that are attributable to a given training image are recorded as microtransactions to the distributed ledger in association with the given training image.
16. The system of claim 15 , wherein the machine learning model is a neural network.
17. The system of claim 15 , wherein the truth data for a given training image comprises diagnostic data associated with the given training image.
18. The system of claim 17 , wherein the prediction comprises a diagnosis of one or more medical conditions.
19. The system of claim 17 , wherein the given training image is an image of a patient, and wherein the diagnostic data comprises one or more medical conditions of the patient at a time when the image was captured.
20. The system of claim 15 , wherein each respective modification of the plurality of modifications made to the machine learning model is registered on the distributed ledger, and wherein the respective modification to the training image that resulted in the modification is attributed to the training image that resulted in the respective modification.