Medical Image Processing Using Machine Learning for Diagnostic Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging technologies often require high-cost equipment, increased radiation exposure, use of contrast media with side effects, or prolonged imaging times to achieve images suitable for diagnosis, and even high-resolution images can be unsuitable due to noise or low contrast.
Innovation Solution
A medical image processing apparatus and method that utilizes a machine learning engine to improve image quality by combining original and enhanced images based on specific image regions, generating a composite image displayed according to a ratio, thereby enhancing diagnostic suitability without the need for expensive equipment or increased invasiveness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a high-performance imaging apparatus is purchased to obtain high image quality, then image quality is improved, but investment cost increases
Solution Approach 1:
The patent uses a machine learning engine to generate a synthetic high-quality image that copies the essential diagnostic features of an image obtained from a high-performance imaging apparatus, but based on input from a low-performance apparatus. This allows the system to produce images with high manufacturing precision (image quality) without requiring the expensive high-performance imaging equipment, thus reducing investment cost.
2Manufacturing precision
If the amount of radiation is increased to obtain an image with less noise, then noise is reduced, but radiation exposure to the subject increases
Solution Approach 1:
The patent replaces the mechanical/physical approach of increasing radiation exposure to reduce noise with an information-processing approach using a machine learning engine. The engine processes the noisy image to generate a low-noise version, substituting the harmful physical method (increased radiation) with a computational method that achieves the same noise reduction goal without exposing the subject to additional harmful factors.
3Manufacturing precision
If a contrast medium is used to obtain an image in which a site is enhanced, then site enhancement is improved, but side effects increase
Solution Approach 1:
The patent substitutes the use of contrast media (chemical method with side effects) with a machine learning-based image processing method. The machine learning engine analyzes the input image to enhance the desired site or lesion without requiring the subject to ingest or receive contrast medium, thereby achieving site enhancement while avoiding the harmful side effects associated with contrast media.
4Manufacturing precision
If the imaging region is widened or spatial resolution is increased to obtain high quality images, then image quality is improved, but imaging time becomes long
Solution Approach 1:
The patent performs preliminary action by using a machine learning engine to process and enhance the image after acquisition. Instead of spending extended time during the imaging process to capture high-quality data, the system quickly acquires the image and then applies computational enhancement to achieve high manufacturing precision (image quality) without the time penalty that would otherwise be required during the actual imaging process.
5Manufacturing precision
If imaging is performed multiple times to obtain high quality images, then image quality is improved, but the time required for imaging increases
Solution Approach 1:
The patent merges multiple approaches by combining the output of a low-performance imaging apparatus with the processing capabilities of a machine learning engine. This integration allows the system to achieve image quality comparable to multiple imaging acquisitions without actually performing multiple scans, thereby reducing the loss of time while maintaining manufacturing precision through the synergistic combination of hardware and software processing.
Data Source
AI summary
A medical image processing apparatus includes: an obtaining unit configured to obtain a first image that is a medical image of a predetermined site of a subject; an image quality improving unit configured to generate, from the first image, a second image in which the image quality is improved compared to the first image by using an image quality improving engine that includes a machine learning engine; and a display controlling unit configured to cause a display unit to display a composite image obtained by combining the first image and the second image according to a ratio obtained using information relating to at least a partial region of the first image.