Hidden Hotspot Detection in Medical MIP Imaging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In medical imaging, particularly with Maximum Intensity Projection (MIP) images, significant pathologies like tumors can be 'hidden' if they do not have the highest voxel value along the lines of projection, leading to incomplete visualization and requiring additional detailed reviews of Multi-Planar Reconstructions (MPRs), which is time-consuming and costly.

Innovation Solution

A method to identify regions of interest by generating an intensity projection image, processing the data to find maxima, comparing them with the projection image, and highlighting maxima not represented in the MIP, using criteria like minimum intensity thresholds or local intensity gradients to ensure detection of 'hidden' hotspots.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If Maximum Intensity Projection (MIP) imaging is used to visualize medical images, then rapid localization of potential lesions is enabled, but hidden hotspots may not be visualized if they do not contain the highest voxel value along projection rays

Engineering Contradiction:
Improverapid localization speedVSAvoidcompleteness of hotspot detection
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system performs preliminary processing by generating an intensity projection image from the medical imaging data before the actual analysis. This preliminary step creates a visual representation that highlights potential hotspots, enabling rapid localization while ensuring completeness by identifying hidden hotspots through subsequent comparison with the projection image.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If detailed review of Multi-Planar Reconstructions (MPRs) is performed to detect hidden hotspots, then complete pathology detection is achieved, but time and costs increase

Engineering Contradiction:
Improvecompleteness of pathology detectionVSAvoidreview time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The intensity projection image serves as an intermediary between the raw medical imaging data and the final analysis. It provides a visual summary that highlights potential hotspots, allowing practitioners to quickly identify regions of interest without needing to perform detailed reviews of all MPRs, thus reducing time while maintaining detection completeness.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Ease of operation

If MIP images are used for initial overview, then quick visualization is achieved, but interaction features with MPRs are not available for MIP-hidden hotspots

Engineering Contradiction:
Improveease of initial overviewVSAvoidinteraction capability with MPRs
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system makes the intensity projection image multi-functional by enabling it to serve both as a quick visualization tool and as an interactive interface. Practitioners can click on hotspots in the intensity projection image to center crosshairs in MPRs or segment them, providing interaction capabilities while maintaining the ease of initial overview.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS9117141B2Method and apparatus for identifying regions of interest in medical imaging data
Publication Date: 2015.08.25 SIEMENS MEDICAL SOLUTIONS USA INC
  • US9117141B2 patent drawing
  • US9117141B2 patent drawing
  • US9117141B2 patent drawing

AI summary

In a method and apparatus for identifying a region of interest in medical imaging data of a subject is described, an intensity projection image is generated from the medical imaging data. The medical imaging data is then processed to find one or more maxima in the medical imaging data. The found maxima are compared with the intensity projection image, and one of the maxima which is not represented in the intensity projection image is identified.