AI Foreign Object Detection for MRI Safety

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for determining the presence of implants or foreign objects in patients before magnetic exposure, such as MRI scans, are unreliable and can lead to serious injuries or deaths due to their dependence on patient memory and additional costly screenings, which may be time-consuming and prone to errors.

Innovation Solution

A system utilizing an AI/ML engine to analyze medical imaging and electronic medical records to detect implants or foreign objects, providing a confidence level or probability of their presence, and assigning an MR Conditional categorization, which can be verified by technologists for accurate pre-exam screening.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional screening methods (patient questionnaires and memory-based evaluation) are used to detect implants or foreign objects before magnetic exposure, then the screening process is simple and quick, but the reliability and accuracy of detection are low, leading to potential serious injuries or deaths

Engineering Contradiction:
Improvedetection accuracyVSAvoidscreening process complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces an AI/ML engine as an intermediary between the patient screening process and the final determination of implant presence. This engine analyzes medical imaging data and electronic health records to detect implants or foreign objects, serving as a mediator that enhances detection accuracy while maintaining a structured screening workflow. The AI system processes complex data patterns that human reviewers might miss, thereby improving reliability without requiring complete redesign of the screening process

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical system of human memory and manual questionnaire completion with an automated AI/ML-based detection system. Instead of relying on patients to accurately recall and report implants, the system uses machine learning algorithms to analyze medical imaging data and automatically detect the presence of implants or foreign objects. This substitution significantly improves detection accuracy while reducing the complexity of patient participation in the screening process

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If additional medical screening (x-rays, CT, metal detection) is performed to check for implants and foreign objects, then the detection accuracy improves, but the time required for screening increases and the process becomes more expensive

Engineering Contradiction:
Improvedetection accuracyVSAvoidscreening time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by using the AI/ML engine to analyze existing medical imaging data and electronic health records before the actual MRI procedure. The system proactively identifies potential implants or foreign objects in advance, allowing the screening process to be completed more efficiently. By performing this analysis beforehand using automated algorithms, the system reduces the need for time-consuming additional screenings during the patient visit while maintaining high detection accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses copies of existing medical data (medical imaging data and electronic health records) to perform the implant detection analysis. Instead of requiring patients to undergo new physical screenings, the AI system analyzes copies of previously collected medical information. This approach maintains detection accuracy while significantly reducing the time and cost associated with additional physical screenings, as the analysis is performed on existing digital records rather than requiring new imaging procedures

Inventive Principle:
Principle #26Copying

3Productivity

If patient memory and self-reporting are relied upon for screening, then the screening process is fast and requires minimal resources, but the accuracy is compromised due to patient misunderstanding, forgetfulness, or unconsciousness

Engineering Contradiction:
Improvescreening efficiencyVSAvoiddetection accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent implements self-service by enabling the medical imaging data and electronic health records to 'speak for themselves' through automated AI analysis. The system allows the data to reveal the presence of implants or foreign objects without requiring patient interpretation or recall. The AI engine independently evaluates the medical records and imaging data, eliminating the need for patients to understand or remember their implant status, thus maintaining high screening efficiency while dramatically improving detection accuracy

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent incorporates feedback mechanisms where the AI/ML engine continuously learns from the analysis of medical imaging data and clinical outcomes. The system uses feedback from detected cases to improve its detection algorithms, ensuring that it accurately identifies implants and foreign objects while maintaining efficient screening processes. This feedback loop allows the system to refine its precision over time without compromising the speed and efficiency of the screening process

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS20240304308A1Foreign object detection for magnetic exposure
Publication Date: 2024.09.12 OPTUM INC
  • US20240304308A1 patent drawing
  • US20240304308A1 patent drawing
  • US20240304308A1 patent drawing

AI summary

A system and method for screening for implants and/or foreign objects prior to undergoing magnetic exposure (e.g., an MRI) utilizing artificial intelligence (AI) analysis of medical imaging information and/or medical records information.