Hepatic Fat Assessment Using Automated Liver-Spleen CT Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing hepatic steatosis, such as liver biopsies and MRI PDFF, are invasive or not widely available, while imaging techniques like CT scans face challenges in accurate manual delineation of regions of interest and high inter-reader variability, especially in low-dose non-contrast CTs.
Innovation Solution
A deep learning pipeline that automatically segments liver and spleen data from digitized images to determine a liver-to-spleen attenuation ratio, reducing inter-reader variability and enabling accurate, reproducible hepatic steatosis determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual delineation of regions of interest is used in CT scans, then flexibility in analysis is maintained, but inter-reader variability increases and measurement precision deteriorates
Solution Approach 1:
The system performs automated segmentation and hepatic steatosis determination without requiring manual intervention. The deep learning model independently identifies liver and spleen regions, calculates attenuation values, and generates diagnostic reports, eliminating inter-reader variability while maintaining measurement precision.
Solution Approach 2:
The patent replaces manual mechanical delineation methods with an automated deep learning-based image processing system. The neural network model automatically segments organs and calculates attenuation ratios, substituting human operators with an algorithmic system that provides consistent, reproducible measurements.
2Measurement precision
If automated deep learning segmentation is implemented, then inter-reader variability is reduced and measurement precision improves, but device complexity increases
Solution Approach 1:
The deep learning pipeline is divided into distinct modular components: liver segmentation, spleen segmentation, attenuation calculation, and diagnostic interpretation. Each module performs a specific function, making the complex system manageable and interpretable while maintaining high measurement precision through specialized processing at each stage.
Solution Approach 2:
The patent introduces automated segmentation maps as intermediary outputs between raw imaging data and final diagnostic conclusions. These segmentation maps serve as intermediate representations that facilitate accurate attenuation ratio calculation and can be visually verified, bridging the gap between complex algorithmic processing and clinical interpretation.
3Productivity
If rapid automated assessment is provided, then productivity increases, but thoroughness of analysis may be compromised
Solution Approach 1:
The deep learning model performs preliminary automated segmentation and attenuation calculation immediately upon receiving imaging data, providing rapid initial assessment. This preliminary action is supplemented by optional quality control checks and visual verification steps, ensuring that speed does not compromise diagnostic reliability.
Solution Approach 2:
The system incorporates feedback mechanisms where segmentation results and attenuation ratios are automatically validated against expected physiological ranges. If anomalies are detected, the system can trigger re-evaluation or alert clinicians for manual review, maintaining diagnostic accuracy while preserving rapid processing for routine cases.
Data Source
AI summary
The present disclosure relates to a method. The method includes accessing automatically segmented liver data and automatically segmented spleen data from a patient. The automatically segmented liver data is used to determine a liver attenuation and the automatically segmented spleen data is used to determine a spleen attenuation. A liver-to-spleen attenuation ratio is determined from the liver attenuation and the spleen attenuation. A hepatic steatosis determination is made from the liver-to-spleen attenuation ratio.