Methods and apparatus for self-learning clinical decision support

Patent metadata

Field	Value
Publication number	US-10629305-B2
Application number	US-201715808536-A
Country	US
Kind code	B2
Filing date	Nov 9, 2017
Priority date	Nov 9, 2017
Publication date	Apr 21, 2020
Grant date	Apr 21, 2020

Abstract

Apparatus, systems, and methods for computer-aided detection are disclosed and described. An example apparatus includes at least one processor and a memory. The example memory includes instructions which, when executed, cause the at least one processor to at least: associate first patient data and first outcome data according to a set of association rules; train a processing model using machine learning and the associated first patient data and first outcome data; generate a computer-aided decision processing algorithm using the processing model; update the computer-aided decision processing algorithm based on at least one of second patient data or second outcome data received from a cloud infrastructure; and deploy the updated computer-aided decision processing algorithm to be applied to third patient data.

First claim

What is claimed is:

1. A method comprising:
associating, using at least one processor, first patient data and first outcome data according to a set of association rules, the set of association rules to organize data retrieved from a cloud infrastructure according to at least one of patient, patient type, or location to form a truthed database;
training, using the at least one processor, a processing model using machine learning to build the processing model using a machine learning network and the associated first patient data and first outcome data from the truthed database;
generating, using the at least one processor, a computer-aided decision processing algorithm implemented using the processing model;
dynamically updating, using the at least one processor, the computer-aided decision processing algorithm based on at least one of second patient data or second outcome data related to a patient characteristic of a target patient, the at least one of second patient data or second outcome data made available via the cloud infrastructure to combine with the truthed database to train an updated processing model; and
deploying, using the at least one processor, the updated computer-aided decision processing algorithm to be applied to third patient data using the updated processing model of the updated computer-aided decision processing algorithm tailored to the target patient associated with the third patient data based on at least the patient characteristic of the target patient.

2. The method of claim 1 , wherein the at least one processor includes an association processor and a learning processor.

3. The method of claim 1 , wherein the first patient data includes first image data and the second patient data includes second image data.

4. The method of claim 1 , wherein the at least one of second patient data or second outcome data is provided periodically by the cloud infrastructure after the at least one of second patient data or second outcome data is acquired and processed by the cloud infrastructure.

5. The method of claim 1 , wherein the processing model includes an artificial neural network.

6. The method of claim 1 , wherein the first patient data and the first outcome data are associated to provide truthed data when an analysis of the first patient data is confirmed by the first outcome data.

7. The method of claim 1 , wherein associating further includes associating first demographic data and first pathology result data with the first patient data and the first outcome data.

8. A non-transitory computer readable medium including instructions which, when executed, cause at least one processor to at least:
associate first patient data and first outcome data according to a set of association rules, the set of association rules to organize data retrieved from a cloud infrastructure according to at least one of patient, patient type, or location to form a truthed database;
train a processing model using machine learning to build the processing model using a machine learning network and the associated first patient data and first outcome data from the truthed database;
generate a computer-aided decision processing algorithm implemented using the processing model;
dynamically update the computer-aided decision processing algorithm based on at least one of second patient data or second outcome data related to a patient characteristic of a target patient, the at least one of second patient data or second outcome data made available via the cloud infrastructure to combine with the truthed database to train an updated processing model; and
deploy the updated computer-aided decision processing algorithm to be applied to third patient data using the updated processing model of the updated computer-aided decision processing algorithm tailored to the target patient associated with the third patient data based on at least the patient characteristic of the target patient.

9. The computer readable medium of claim 8 , wherein the first patient data includes first image data and the second patient data includes second image data.

10. The computer readable medium of claim 8 , wherein the at least one of second patient data or second outcome data is provided periodically by the cloud infrastructure after the at least one of second patient data or second outcome data is acquired and processed by the cloud infrastructure.

11. The computer readable medium of claim 8 , wherein the processing model includes an artificial neural network.

12. The computer readable medium of claim 8 , wherein the first patient data and the first outcome data are associated to provide truthed data when an analysis of the first patient data is confirmed by the first outcome data.

13. The computer readable medium of claim 8 , wherein associating further includes associating first demographic data and first pathology result data with the first patient data and the first outcome data.

14. An apparatus including at least one processor and a memory, the memory including instructions which, when executed, cause the at least one processor to at least:
associate first patient data and first outcome data according to a set of association rules, the set of association rules to organize data retrieved from a cloud infrastructure according to at least one of patient, patient type, or location to form a truthed database;
train a processing model using machine learning to build the processing model using a machine learning network and the associated first patient data and first outcome data from the truthed database;
generate a computer-aided decision processing algorithm implemented using the processing model;
dynamically update the computer-aided decision processing algorithm based on at least one of second patient data or second outcome data related to a patient characteristic of a target patient, the at least one of second patient data or second outcome data made available via the cloud infrastructure to combine with the truthed database to train an updated processing model; and
deploy the updated computer-aided decision processing algorithm to be applied to third patient data using the updated processing model of the updated computer-aided decision processing algorithm tailored to the target patient associated with the third patient data based on at least the patient characteristic of the target patient.

15. The apparatus of claim 14 , wherein the at least one processor includes an association processor and a learning processor.

16. The apparatus of claim 14 , wherein the first patient data includes first image data and the second patient data includes second image data.

17. The apparatus of claim 14 , wherein the at least one of second patient data or second outcome data is provided periodically by the cloud infrastructure after the at least one of second patient data or second outcome data is acquired and processed by the cloud infrastructure.

18. The apparatus of claim 14 , wherein the processing model includes an artificial neural network.

19. The apparatus of claim 14 , wherein the first patient data and the first outcome data are associated to provide truthed data when an analysis of the first patient data is confirmed by the first outcome data.

20. The apparatus of claim 14 , wherein associating further includes associating first demographic data and first pathology result data with the first patient data and the first outcome data.

Assignees

• Gen Electric

Inventors

• Muller Serge
• Crasemann Axel
• Blumenthal Marco
• Bernard Sylvain
• Iordache Razvan

Classifications

• G16H50/20 Primary

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title

• G06N3/02

  Neural networks · CPC title