CT Image Texture Analysis for Noninvasive HCC Grade Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for pre-operatively assessing the histological grade of Hepatocellular Carcinoma (HCC) are limited by the risk of tumor seeding during biopsy and the inability to reflect heterogeneous histology, necessitating a noninvasive and accurate prediction of tumor grade.
Innovation Solution
A method utilizing Computed Tomography (CT) images, where filters are applied through convolution operations to produce response images, with Independent Subspace Analysis (ISA) features being computed and classified using Random Forest regression to determine the histological grade of HCC tumors pre-operatively and non-invasively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If biopsy is performed for pre-operative assessment of HCC histological grade, then tumor grade information can be obtained, but risk of tumor seeding along biopsy tract occurs
Solution Approach 1:
The patent replaces the mechanical biopsy procedure with a non-invasive CT imaging-based computational method. Image filters are applied to CT images to extract texture features that correlate with histological grade, eliminating the need for physical tissue sampling and thus avoiding tumor seeding risk while still providing accurate tumor grade assessment
Solution Approach 2:
The patent creates a virtual copy of the tumor's histological characteristics through image processing. By applying trained filters to CT images and extracting texture features, the system reproduces the diagnostic information normally obtained from biopsy without physical contact, thereby preventing tumor seeding while maintaining measurement precision
2Measurement precision
If biopsy is performed for HCC grading, then tumor grade can be determined, but sampling errors occur due to heterogeneous histology
Solution Approach 1:
The patent merges information from multiple image filters and multiple CT image phases (arterial, portal venous, delayed) to create a comprehensive assessment of the entire tumor. This combines diverse texture features and temporal information to capture histological heterogeneity across the whole tumor volume, eliminating sampling errors while maintaining accurate grade determination
Solution Approach 2:
The patent transitions from 2D biopsy samples to 3D volumetric CT image analysis. By processing multi-phase CT scans through multiple filters and synthesizing information across different time points and spatial dimensions, the system captures the full heterogeneity of the tumor without the sampling limitations of traditional biopsy
3Ease of operation
If CT imaging is used for HCC identification, then tumor localization is achieved, but HCC tumor grade prediction capability is lacking
Solution Approach 1:
The patent enhances the CT imaging system to perform multiple functions: tumor detection, localization, characterization, and grade prediction. By integrating trained image filters and a classification system that processes texture features from multi-phase images, the same CT scan that identifies the tumor also provides histological grade information, eliminating the need for separate diagnostic procedures
Data Source
AI summary
A method of determining the histological grade of Hepatocellular Carcinoma (HCC) including: acquiring a Computed Tomography (CT) image of a person including an HCC tumor; delineating the HCC tumor; and assigning a histological grade to the HCC tumor, wherein assigning the histological grade to the HCC tumor includes: applying a plurality of filters to the HCC tumor, wherein each of the filters produces a corresponding response image and, for each of the filters, a convolution operation is performed on the filter and the CT image to produce the response image corresponding to that filter; computing an average response of the HCC tumor in each of the response images and recording each of the average responses as an Independent Subspace Analysis (ISA) feature; and determining the histological grade of the HCC tumor based on the ISA features by using a classifier.