MEDICAL EVALUATION MACHINE LEARNING WORKFLOWS AND PROCESSES

• US 20160350919A1
• Filed: 05/31/2016
• Published: 12/01/2016
• Est. Priority Date: 06/01/2015
• Status: Active Grant

• Current Assignee: Virtual Radiologic Corporation (Mednax Incorporated)
• Original Assignee: Virtual Radiologic Corporation (Mednax Incorporated)
• Inventors: Strong, Benjamin, Werb, Shannon, Steigauf, Wade J.
• Granted Patent: US 9,846,938 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 18/2148 characterised by the process organisation or structure, e.g. boosting cascade
• CPC Class Codes: G06F 18/2178 based on feedback of a supervisor
• CPC Class Codes: G06F 18/2413 based on distances to training or reference patterns
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06T 2207/10081 Computed x-ray tomography [CT]
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/10116 X-ray image
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/20084 Artificial neural networks [ANN]
• CPC Class Codes: G06T 7/0014 using an image reference approach
• CPC Class Codes: G06V 10/451 with interaction between the filter responses, e.g. cortical complex cells
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects
• CPC Class Codes: G06V 10/82 using neural networks
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing
• CPC Class Codes: G16H 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Systems and methods for processing electronic imaging data obtained from medical imaging procedures are disclosed herein. Some embodiments relate to data processing mechanisms for medical imaging and diagnostic workflows involving the use of machine learning techniques such as deep learning, artificial neural networks, and related algorithms that perform machine recognition of specific features and conditions in imaging data. In an example, a deep learning model is selected for automated image recognition of a particular medical condition on image data, and applied to the image data to recognize characteristics of the particular medical condition. Based on the characteristics recognized by the automated image recognition on the image data, an electronic workflow for performing a diagnostic evaluation of the medical imaging study may be modified, updated, or prioritized.

Claims

1. A method of electronic processing for data in a medical evaluation workflow, comprising a plurality of electronic operations executed with a processor and memory of a computing device, with the plurality of electronic operations comprising:

obtaining image data associated with a medical imaging study;

obtaining non-image data associated with the medical imaging study;

selecting a deep learning model to apply automated image recognition of a particular medical condition on the image data, wherein the deep learning model is selected based on information indicated in metadata of the image data and information indicated in a data field of the non-image data;

recognizing characteristics of the particular medical condition from image content of the image data, wherein the characteristics of the particular medical condition are recognized using automated image recognition with the deep learning model; and

modifying an electronic workflow for performing a diagnostic evaluation of the medical imaging study, wherein the electronic workflow is modified based on the characteristics of the particular medical condition recognized from the image data.

2. The method of claim 1, the electronic operations further comprising:

generating a score that corresponds to a level of recognition of the characteristics of the particular medical condition in the image data, wherein the particular medical condition is identified as being shown or not shown in the image content of the image data based on the score.

3. The method of claim 2,

wherein the characteristics of the particular medical condition recognized in the image data include a positive finding and a negative finding related to the particular medical condition.

4. The method of claim 1, wherein modifying the electronic workflow for performing the diagnostic evaluation of the medical imaging study includes:

determining an assignment for the medical imaging study to a server-managed evaluator worklist corresponding to a computing system of a human evaluator;

identifying that the medical imaging study includes a prioritized medical condition based on the characteristics of the particular medical condition recognized from the image data; and

prioritizing the assignment for the medical imaging study in the server-managed evaluator worklist, wherein the prioritizing designates the assignment for the medical imaging study over a prior assignment for another medical imaging study that does not include the prioritized medical condition.

5. The method of claim 4, the electronic operations further comprising:

adding the assignment for the medical imaging study to an additional server-managed evaluator worklist based on the identifying of the prioritized medical condition, wherein the additional server-managed evaluator worklist corresponds to another computing system of another human evaluator.

6. The method of claim 4, the electronic operations further comprising:

prioritizing results and reporting from the computing system of the human evaluator for the medical imaging study.

7. The method of claim 1, the electronic operations further comprising:

unbundling a plurality of series of images within the image data based on an indication of a trauma or other time-sensitive medical condition in the characteristics of the particular medical condition, wherein the plurality of series of images are unbundled into respective groupings based on anatomical regions captured by the plurality of series of images;

wherein modifying the electronic workflow for performing the diagnostic evaluation of the medical imaging study includes assigning and prioritizing the respective groupings within the medical imaging study in a first worklist and a second worklist, wherein the first worklist corresponds to a first evaluator and wherein the second worklist corresponds to a second evaluator.

8. The method of claim 1, further comprising:

training an image data recognition algorithm used in the deep learning model, wherein the training analyzes one or more historical image data sets and report data associated with the one or more historical image data sets to identify the characteristics of the particular medical condition.

9. The method of claim 8, the electronic operations further comprising:

identifying the particular medical condition for training;

identifying the particular medical condition in prior evaluation reports corresponding to a historical set of imaging studies;

identifying findings of the particular medical condition in images of the historical set of imaging studies, wherein the identified findings include positive findings and negative findings relative to the particular medical condition that were determined by a human evaluator; and

training image recognition classifiers of the deep learning model, the training using the findings of the particular medical condition with the images from the historical set of imaging studies.

10. The method of claim 9, the electronic operations further comprising:

de-identifying protected health information from the prior evaluation reports and the images of the historical set of imaging studies;

wherein the prior evaluation reports are produced from respective diagnostic evaluations of images produced for the historical set of imaging studies.

11. The method of claim 8, the electronic operations further comprising:

verifying training of the deep learning model based on results indicated by a human evaluator, wherein the results are produced from the diagnostic evaluation of the medical imaging study.

12. The method of claim 1,

wherein the image data is provided from a radiological imaging procedure,

wherein the non-image data is provided from a radiological imaging order, and

wherein the medical imaging study corresponds to a radiological read request for diagnostic evaluation of the image data by a medical professional evaluator, the radiological read request indicated by the radiological imaging order.

13. A non-transitory device-readable storage medium, the device-readable storage medium including instructions that, when executed by a processor and memory of a computing device, causes the computing device to perform operations that:

identify image data associated with a medical imaging study;

identify non-image data associated with the medical imaging study;

select a deep learning model to apply automated image recognition of a particular medical condition on the image data, wherein the deep learning model is selected based on information indicated in metadata of the image data and information indicated in a data field of the non-image data;

recognize characteristics of the particular medical condition from image content of the image data, wherein the characteristics of the particular medical condition are recognized using automated image recognition with the deep learning model; and

modify an electronic workflow for performing a diagnostic evaluation of the medical imaging study, wherein the electronic workflow is modified based on the characteristics of the particular medical condition recognized from the image data.

14. The non-transitory device-readable storage medium of claim 13, the instructions further to cause the computing device to perform operations that:

determine an assignment for the medical imaging study to a server-managed evaluator worklist corresponding to a computing system of a human evaluator;

identify that the medical imaging study includes a prioritized medical condition based on the characteristics of the particular medical condition recognized from the image data; and

perform prioritization of the assignment for the medical imaging study in the server-managed evaluator worklist, wherein the prioritization designates the assignment for the medical imaging study over a prior assignment for another medical imaging study that does not include the prioritized medical condition.

15. The non-transitory device-readable storage medium of claim 14, wherein prioritization of the assignment for the medical imaging study in the evaluator worklist based on the prioritized medical condition, further causes the computing device to perform operations that:

unbundle a plurality of series of images within the medical imaging study based on an indication of a trauma or other time-sensitive medical condition in the characteristics of the particular medical condition, wherein the plurality of series of images are unbundled into respective groupings based on anatomical regions captured in the plurality of series of images;

wherein modification of the electronic workflow to perform the diagnostic evaluation of the medical imaging study includes assignment and prioritization of the respective groupings within the medical imaging study in a first worklist and a second worklist, wherein the first worklist corresponds to a first evaluator and wherein the second worklist corresponds to a second evaluator.

16. The non-transitory device-readable storage medium of claim 14,

wherein the image data is provided from a radiological imaging procedure,

wherein the non-image data is provided from a radiological imaging order, and

wherein the medical imaging study corresponds to a request for diagnostic evaluation of the image data by a medical professional evaluator, the request indicated by the radiological imaging order.

17. A computing system, comprising:

a processor; and

a memory device comprising instructions stored thereon, which when executed by the processor, configure the processor to perform electronic operations with the computing system that;

extract image data associated with a medical imaging study;

extract non-image data associated with the medical imaging study;

select a deep learning model to apply automated image recognition of a particular medical condition on the image data, wherein the deep learning model is selected based on information indicated in metadata of the image data and information indicated in a data field of the non-image data;

recognize characteristics of the particular medical condition from image content of the image data, wherein the characteristics of the particular medical condition are recognized using automated image recognition with the deep learning model; and

modify an electronic workflow for performing a diagnostic evaluation of the medical imaging study, wherein the electronic workflow is modified based on the characteristics of the particular medical condition recognized from the image data.

18. The computing system of claim 17, the processor further to perform electronic operations with the computing system that:

determine an assignment for the medical imaging study to a server-managed evaluator worklist corresponding to a computing system of a human evaluator;

identify that the medical imaging study includes a prioritized medical condition based on the characteristics of the particular medical condition recognized from the image data; and

perform prioritization of the assignment for the medical imaging study in the server-managed evaluator worklist, wherein the prioritization designates the assignment for the medical imaging study over a prior assignment for another medical imaging study that does not include the prioritized medical condition.

19. The computing system of claim 18, wherein the prioritization of the assignment for the medical imaging study in the evaluator worklist based on the prioritized medical condition, further causes the processor to perform electronic operations that:

unbundle a plurality of series of images within the medical imaging study based on an indication of a trauma or other time-sensitive medical condition in the characteristics of the particular medical condition, wherein the plurality of series of images are unbundled into respective groupings based on anatomical regions captured in the plurality of series of images;

wherein modification of the electronic workflow to perform the diagnostic evaluation of the medical imaging study includes assignment and prioritization of the respective groupings within the medical imaging study in a first worklist and a second worklist, wherein the first worklist corresponds to a first evaluator and wherein the second worklist corresponds to a second evaluator.

20. The computing system of claim 17,

wherein the image data is provided from a radiological imaging procedure,

wherein the non-image data is provided from a radiological imaging order, and

wherein the medical imaging study corresponds to a request for diagnostic evaluation of the image data by a medical professional evaluator, the request indicated by the radiological imaging order.