Lesion Analysis Method Using Segmented Detection Modules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting and measuring lesions in medical images fail to provide sufficient information for disease diagnosis, as they primarily focus on identification rather than processing and presenting lesion data in a form suitable for diagnosis.
Innovation Solution
A method utilizing a computing device that generates probability values and location information for nodules in medical images through pre-processing, detection, and post-processing modules, including neural network-based sub-modules for feature map generation and clustering, to determine the presence, size, and classification of suspicious nodules, and subsequently provides a user interface with diagnostic information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional automated methods are used for detecting lesions, then identification of lesions is achieved, but the information provided is insufficient for disease diagnosis
Solution Approach 1:
The patent segments the lesion analysis process into multiple specialized modules: detection module for locating lesions, measurement module for quantifying lesion characteristics, and classification module for diagnosing disease types. Each module processes specific aspects of lesion information independently, ensuring comprehensive diagnostic data is captured and preserved without information loss.
Solution Approach 2:
The patent transitions from 2D medical images to 3D volumetric analysis by constructing three-dimensional lesion models from multiple two-dimensional slices. This dimensional expansion enables extraction of depth information, volume measurements, and spatial relationships that are critical for accurate disease diagnosis while preserving all diagnostic information.
2Ease of operation
If conventional lesion detection methods are used, then basic identification is achieved, but processing and presenting lesion data in diagnostic form is insufficient
Solution Approach 1:
The patent introduces an information processing layer that acts as an intermediary between raw lesion detection and clinical diagnosis. This layer includes measurement modules that calculate quantitative parameters (volume, density, growth rate) and classification modules that interpret these parameters according to diagnostic criteria, transforming raw data into clinically actionable information.
Solution Approach 2:
The patent transforms qualitative lesion characteristics into quantitative parameters through systematic measurement. Lesion volume, density, margin characteristics, and growth rates are calculated as numerical values that can be objectively compared against diagnostic thresholds, making the information reliable and suitable for clinical decision-making.
3Productivity
If automated detection is implemented, then identification speed is improved, but diagnostic-quality information processing is lacking
Solution Approach 1:
The patent divides the automated analysis system into specialized modules that process different aspects of lesion information in parallel: detection module for rapid localization, measurement module for quantitative analysis, and classification module for diagnostic interpretation. This segmentation enables high-speed processing while preserving comprehensive diagnostic information through distributed specialized computation.
Solution Approach 2:
The patent performs preliminary processing of medical images including normalization, enhancement, and feature extraction before main analysis. This preliminary action prepares the data in advance, enabling faster subsequent processing while ensuring all diagnostic information is preserved and enhanced through preprocessing operations.
Data Source
AI summary
Disclosed is a method for analyzing a lesion based on a medical image performed by a computing device. The method may includes generating, by using a pre-processing module, an input image of a pre-trained detection module from the medical image. The method may include generating, by using the detection module, a probability value regarding a presence of a nodule in at least one region of interest and first location information about the at least one region of interest, based on the input image. The method may include determining, by using a post-processing module, second location information about a suspicious nodule present in the medical image from the first location information, based on the probability value regarding the presence of the nodule.