Two-Phase Disease Diagnosis System Using Four-Channel Input Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional disease diagnosis systems using neural networks are limited in accuracy as they only consider the specific patch of a biometric image and not the surrounding tissue, and are sensitive to color variations due to dyeing reagents, which can lead to incomplete diagnosis results.
Innovation Solution
A two-phase disease diagnosis system that includes a patch neural network receiving four-channel input data (RGB and gray channels) to provide patch-level diagnosis and a slide diagnosis engine that clusters cancerous patches to determine disease presence across the entire slide, using cluster features to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only the specific patch is considered for diagnosis, then the device complexity is reduced, but the measurement precision of disease state is insufficient
Solution Approach 1:
The diagnosis system is segmented into two independent phases: patch-level diagnosis (using CNN to analyze individual patches) and slide-level diagnosis (using clustering algorithms to aggregate patch results). This segmentation allows each component to be optimized independently while achieving comprehensive diagnostic accuracy that considers both specific patches and surrounding tissue context.
2Reliability
If RGB color information is used as input data, then the ease of operation is improved, but the reliability of diagnosis is reduced due to sensitivity to dyeing variations
Solution Approach 1:
The input data parameters are transformed from RGB color space to HSV color space, separating hue (color information) from saturation and value (brightness). This parameter change makes the diagnosis system robust to variations in dyeing intensity and lighting conditions, as the hue channel remains relatively stable while saturation and value may vary. The system processes all three channels but weights the hue channel more heavily for disease detection.
3Measurement precision
If patch-level diagnosis is performed independently, then the productivity is improved, but the measurement precision of overall disease presence is reduced
Solution Approach 1:
The system performs preliminary patch-level diagnosis for all patches in parallel using the trained CNN model, storing the results. Then, in a subsequent slide-level diagnosis phase, clustering algorithms aggregate these pre-computed patch results to determine overall disease presence. This preliminary action allows efficient parallel processing while ensuring comprehensive diagnostic accuracy through the two-phase approach.
Data Source
AI summary
A two-phase disease diagnosis system includes a processor and a storage device storing a neural network, and using a slide including a biometric image and the neural network. The system includes a patch neural network that receives, through an input layer, a given patch segmented in a given size from the slide and outputs patch level diagnosis results indicating whether a disease is present in the patch and a slide diagnosis engine that marks a patch determined to be cancer based on the patch level diagnosis results for each of multiple patches included in the slide and outputs slide level diagnosis results indicating whether a disease is present in the slide based on the marked results. The patch neural network receives, through the input layer, four-channel information including original color information three-channels and a gray channel for the patch.