ROI-Based Medical Image Compression for Diagnostic Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing data size of DICOM images due to enhanced image quality in medical imaging and the uniform application of lossy compression poses challenges in storage capacity and communication network overload, risking the loss of important image information.
Innovation Solution
A medical data processing apparatus determines regions of interest and sets varying compression ratios based on their importance, applying lossless or lossy compression accordingly to maintain image quality and reduce data size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If lossy compression is uniformly applied to all medical images, then data size is reduced and storage capacity is improved, but important image information may be lost and diagnostic accuracy deteriorates
Solution Approach 1:
The patent segments the medical image into multiple regions of interest (ROIs) based on anatomical structures, pathological findings, and diagnostic importance. Each ROI is assigned a different compression ratio, with critical regions using lossless or low-compression ratios and non-critical regions using higher compression ratios. This segmentation approach allows selective compression that preserves diagnostically important information while reducing overall data size.
Solution Approach 2:
The patent applies local quality by assigning different compression qualities to different regions within the same image. Critical regions such as lesions, abnormalities, or anatomically important structures are compressed with higher quality (lower compression ratios), while non-critical background areas use lower quality (higher compression ratios). This ensures that diagnostically important areas maintain their informational integrity.
2Quantity of substance
If lossy compression is applied to all images including those stored for follow-up, then storage capacity is optimized, but the quality necessary for future follow-up diagnosis deteriorates
Solution Approach 1:
The patent performs preliminary identification of regions that will be important for future follow-up diagnosis by analyzing the current image for potential pathological changes, anatomical landmarks, and areas that may require future comparison. These identified regions are marked and assigned appropriate compression ratios that preserve their quality for future diagnostic purposes, even before the actual follow-up occurs.
Solution Approach 2:
The patent dynamically adjusts compression parameters based on the diagnostic importance and follow-up requirements of each region. By changing compression ratios, quality levels, and preservation priorities according to region-specific parameters, the system optimizes both current storage efficiency and future diagnostic reliability without uniform compression.
3Productivity
If high compression ratios are applied to reduce data size, then transmission efficiency is improved, but the accuracy of medical image analysis and diagnosis deteriorates
Solution Approach 1:
The patent segments the image into regions requiring high transmission efficiency and regions requiring high analysis accuracy. By segmenting based on diagnostic importance, the system can apply higher compression ratios to non-critical regions (improving transmission efficiency) while maintaining lower compression ratios for critical regions (preserving analysis accuracy).
Solution Approach 2:
The patent applies different quality levels locally across the image, with critical diagnostic regions maintained at high quality (low compression) and non-critical regions compressed more aggressively. This local quality differentiation allows the overall transmission efficiency to improve while the accuracy of image analysis for critical regions remains preserved.
Data Source
AI summary
According to one embodiment, a medical data processing apparatus includes processing circuitry. The processing circuitry determines at least one region of interest for performing processing on a medical image and a degree of importance of the region of interest. The processing circuitry calculates a compression ratio for each of the regions of interest according to the degree of importance.