Automated Bullseye Plot Generation via Landmark Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual generation of bullseye plots for cardiac analysis is time-consuming and prone to human error, limiting the efficiency and accuracy of cardiac diagnosis in medical imaging.
Innovation Solution
A system utilizing a machine learning technique to automatically generate bullseye plots by obtaining and segmenting multiple slice images of the heart, identifying landmarks using a landmark detection network, and segmenting the myocardium into anatomical regions, thereby creating a bullseye plot with parameter values indicative of physiological conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual methods are used to generate bullseye plots, then flexibility and adaptability are maintained, but time consumption and human error increase
Solution Approach 1:
The system performs automatic bullseye plot generation using machine learning models that independently identify landmarks and segment myocardium without requiring manual intervention. The landmark detection network and cardiac segmentation network autonomously process slice images to generate accurate bullseye plots, eliminating human error while maintaining high productivity.
Solution Approach 2:
The patent replaces manual mechanical operations with automated computational systems. Instead of manually identifying landmarks and segmenting myocardium, the system uses deep learning models (landmark detection network and cardiac segmentation network) to automatically process images and generate bullseye plots, significantly improving both speed and accuracy.
2Loss of time
If manual generation is used, then ease of operation is maintained for simple cases, but time consumption and subjectivity increase
Solution Approach 1:
The system divides the complex task of bullseye plot generation into distinct segments: landmark detection using a specialized detection network, myocardium segmentation using a cardiac segmentation network, and final plot generation. This modular approach manages complexity while enabling rapid automated processing of multiple slice images.
Solution Approach 2:
The system performs preliminary actions by pre-training the landmark detection network and cardiac segmentation network on extensive datasets before deployment. This preliminary training enables the networks to rapidly and accurately process new slice images without requiring manual intervention, reducing generation time while managing complexity through prior preparation.
3Reliability
If automated methods are implemented, then productivity and accuracy are improved, but measurement precision requirements increase
Solution Approach 1:
The landmark detection network incorporates feedback mechanisms where the detected landmarks are used to guide the cardiac segmentation network, which in turn refines the overall bullseye plot generation. This feedback loop ensures that high precision in landmark detection translates to accurate myocardium segmentation and reliable objective results.
Solution Approach 2:
The patent introduces distance offset maps as an intermediary representation between the input slice images and the final landmark positions. The landmark detection network first generates distance offset maps that encode spatial relationships, which then serve as intermediaries for precise landmark identification. This intermediary step enhances measurement precision while maintaining objectivity.
Data Source
AI summary
A system for generating a bullseye plot of a heart of a subject is provided. The system may obtain multiple slice images in a plurality of groups, wherein each group corresponds to one of a plurality of sections of the heart and includes at least one slice image of the corresponding section, and each slice image includes part of the right ventricle, part of the left ventricle, and part of the myocardium. The system may also identify at least one landmark associated with the left ventricle by applying a landmark detection network in each of the slice images. The system may further generate the bullseye plot of the heart based on the at least one landmark identified in each of the multiple slice images, wherein the bullseye plot includes a plurality of sectors, each of which represents an anatomical region of the myocardium in one of the plurality of sections.