Data aggregation, integration and analysis system and related devices and methods
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11004565-B2
Publication Date
2021-05-11
Expiration Date
Abstract
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.
Core Innovation
The disclosed diagnostic-data system records image data, image-specific ground truth, machine-learning training parameters, and model modifications on a ledger. Each data-linked learning update is treated as a microtransaction. The records associate learning and subsequent model changes with contributing images, while model performance is assessed on a test data set as part of the attribution process.
The supplied partial content does not include the background section, so the problem being solved is not explicitly described in patent. The content identifies AUC, sensitivity, and specificity as performance metrics and depicts an imaging device, server, learning platform, blockchain, and data collection, aggregation, training, and attribution flow for the diagnostic-data system.
Claims Coverage
The claims recite two inventive features across two independent claims, claim 1 directed to a non-transitory computer-readable medium and claim 11 directed to a method.
Ledger-based recording and image attribution
Claim 1 provides instructions and claim 11 recites a method to register a pool of images and image-specific truth data sets on a ledger, register training parameters to the images, and register each machine-learning algorithm modification while attributing it to the image that resulted in the modification.
Performance assessment and image attribution
Claim 1 provides instructions and claim 11 recites a method to assess machine-learning algorithm performance with each modification on a test data set and attribute any change to the image that resulted in the change.
Both independent claims cover ledger recording of images, image-specific truth data, training information, and algorithm modifications, together with assessment of performance changes and attribution of those changes to images.
Stated Advantages
Not explicitly described in patent.
Documented Applications
A diagnostic-data system for recording image data, image-specific ground truth, machine-learning training parameters, and model modifications, and attributing performance changes to contributing images.
Interested in licensing this patent?