Raw Medical Imaging Data Processing for Diagnostic Timeliness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging diagnostic systems face challenges in timely image reconstruction and reliability due to high computation times and introduction of artifacts, which can hinder immediate clinical decision-making, especially in emergencies.
Innovation Solution
The system processes raw medical imaging data, such as MRI k-space or CT sinogram data, using a medical imaging diagnostic controller that includes dimension reduction and artificial intelligence engines to render diagnostic assessments directly from the raw data, eliminating the need for intermediate image reconstruction and reducing computation time while minimizing artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image reconstruction is performed on raw medical imaging data, then diagnostic reliability is improved, but computation time increases substantially
Solution Approach 1:
The patent extracts and removes the image reconstruction step from the traditional diagnostic workflow. Instead of reconstructing raw data into images for diagnosis, the system directly processes raw medical imaging data (sinograms, k-space data) through machine learning models to generate diagnostic assessments, thereby eliminating the time-consuming reconstruction process while maintaining diagnostic accuracy
Solution Approach 2:
The patent inverts the conventional diagnostic approach by not converting raw data to images for interpretation. Instead, it feeds raw data directly into AI models that have been trained to diagnose conditions from the original acquired data format, reversing the traditional data transformation pipeline
2Reliability
If image reconstruction is performed on raw medical imaging data, then diagnostic assessment is enabled, but artifacts are introduced into the image
Solution Approach 1:
The patent removes the image reconstruction process entirely from the diagnostic pathway. By directly analyzing raw medical imaging data through machine learning models, the system eliminates the source of reconstruction artifacts (such as streak artifacts, Gibbs ringing, and zipper interference) that contaminate reconstructed images and can lead to misdiagnosis
Solution Approach 2:
The patent prevents artifact formation by never performing the reconstruction operation that creates them. The machine learning models are trained directly on raw data, so the harmful transformation that introduces artifacts is preemptively avoided
3Ease of operation
If image reconstruction is performed on raw medical imaging data, then medical image is generated for interpretation, but dynamic range is reduced in display
Solution Approach 1:
The patent extracts and eliminates the image generation step that compresses the dynamic range. By performing diagnosis directly on raw data, the system preserves the full dynamic range information present in the original acquired data without the loss inherent in reconstruction and display processes
Data Source
Figure 1A~1B
Figure 2~3
Figure 4A~4C
AI summary
Various embodiments of the present disclosure are directed to a raw diagnostic machine for a medical diagnosis of raw medical imaging data generated by a medical imaging machine as opposed to a medical diagnosis of a medical image conventionally reconstructed from the raw medical imaging data. In operation, the raw diagnostic engine includes a medical imaging diagnostic controller implementing a dimension reduction pre-processor for selecting or extracting one or more dimension reduced feature vectors from the raw medical imaging data, and further implementing a raw diagnostic artificial intelligence engine for rendering a diagnostic assessment of the raw medical imaging data as represented by the dimension reduced feature vector(s). The medical imaging diagnostic controller may further control a communication of the diagnostic assessment of the raw medical imaging data (e.g., a display, a printing, an emailing, a texting, etc.).