Hybrid Cloud Local Cognitive Medical Image Measurement

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Manual measurement of anatomical features in medical images is tedious and time-consuming, and existing cloud-based automation solutions often suffer from long turnaround times due to computational limitations.

Innovation Solution

Implementing a hybrid cloud/local cognitive measurement system where pre-trained parameters are received from a remote server to instantiate a local cognitive system for evaluating medical images, allowing for automated or semi-automated measurements with editable boundaries, and providing feedback for further training, utilizing various types of cognitive systems such as artificial neural networks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If cloud-based automated measurement is used, then measurement automation is achieved, but turnaround time becomes too long

Engineering Contradiction:

Improvemeasurement automationVSAvoidturnaround time

Core Design Contradiction:

Extent of automationVSLoss of time

Solution Approach 1:

The system segments the measurement automation task between cloud-based cognitive learning (for model training and parameter updates) and local execution (for actual measurements). The cognitive system is divided into a remote server component that handles learning and a local client component that handles inference, allowing automated measurements to run locally without cloud latency.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary action by pre-training the cognitive system in the cloud and downloading pre-trained parameters to the local system before actual measurements are needed. This preliminary cloud-based training enables fast local execution without requiring cloud connectivity during measurement operations.

Inventive Principle:

Principle #10Preliminary action

2Loss of time

If fully local cognitive system is used, then fast measurements are achieved, but learning capability is limited

Engineering Contradiction:

Improvemeasurement speedVSAvoidlearning capability

Core Design Contradiction:

Loss of timeVSAdaptability or versatility

Solution Approach 1:

The system uses an intermediary approach where pre-trained parameters serve as the bridge between cloud-based learning and local execution. The cognitive system architecture acts as an intermediary, with the local system executing measurements using parameters that were trained in the cloud, and optionally sending feedback results back for continuous learning improvement.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If manual measurement is used, then measurement accuracy can be verified, but time consumption increases

Engineering Contradiction:

Improvemeasurement accuracyVSAvoidtime consumption

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The cognitive measurement system performs self-service by automatically executing measurements and generating results without requiring manual verification for every case. The system serves itself by using the pre-trained cognitive model to perform measurements autonomously, freeing radiologists from tedious manual measurement tasks while maintaining accuracy through the trained model's capabilities.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11080846B2Hybrid cloud-based measurement automation in medical imagery

Publication Date: 2021.08.03 MERATIVE US LP

AI summary

Measurement of medical images as a hybrid cloud service is provided. In various embodiments, pre-trained parameters are received at a client from a remote server. A local cognitive system is instantiated using the pre-trained parameters. The cognitive system is applied to evaluate a medical image. A result is sent to the remote server for training of a remote cognitive system.