AI MRI to CT Conversion for MRgFUS Treatment Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The use of CT scans in magnetic resonance-guided focused ultrasound (MRgFUS) treatments poses a temporal, economical, and radiation exposure burden, particularly for pregnant or elderly patients, due to the need for additional imaging that can cause cell damage and complications.
Innovation Solution
A method using a trainable artificial neural network model to convert MRI images into CT images, eliminating the need for separate CT scans by generating precise CT images based on MRI data through preprocessing, training, and adversarial training between generators and discriminators, enabling accurate skull factor and acoustic property information acquisition for ultrasound treatment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CT scans are used to acquire skull factor information and acoustic parameters for MRgFUS treatment, then the accuracy of ultrasound treatment is improved, but the patient is exposed to radiation and experiences increased temporal and economical burden
Solution Approach 1:
The patent creates a synthetic CT image that copies the essential information from real CT scans (skull factors, acoustic parameters, density, speed of sound) without requiring actual radiation exposure. The GAN-generated synthetic CT serves as a safe replica that contains all necessary treatment planning data.
Solution Approach 2:
The patent introduces MRI images as an intermediary that bridges the gap between available non-radiation imaging and required CT information. By using MRI as input to generate synthetic CT, the system mediates between safe imaging and accurate treatment planning.
2Manufacturing precision
If separate CT scans are performed to acquire acoustic property information, then the precision of ultrasound treatment planning is improved, but the treatment process complexity and time consumption increase
Solution Approach 1:
The patent merges the CT scanning step with the MRI imaging process by generating synthetic CT information from MRI data. This consolidation eliminates the need for separate CT equipment and scanning procedures, reducing overall process complexity while maintaining treatment planning precision.
Solution Approach 2:
The patent makes the MRI imaging process multi-functional by enabling it to serve both as the primary diagnostic imaging modality and as the source for generating CT-equivalent information. This universal approach eliminates the need for specialized CT equipment in the treatment workflow.
3Measurement precision
If multiple imaging modalities (MRI and CT) are used for MRgFUS treatment, then the accuracy of acoustic parameter acquisition is improved, but the temporal burden and cost increase
Solution Approach 1:
The patent performs preliminary action by generating the synthetic CT information from MRI data before the actual ultrasound treatment planning begins. This advance preparation eliminates the need for time-consuming separate CT scanning and processing steps during the treatment workflow.
Solution Approach 2:
The patent ensures continuity of useful action by maintaining the treatment planning process as a continuous workflow using only MRI data. The synthetic CT generation is seamlessly integrated into the MRI-based workflow, eliminating interruptions and waiting times associated with separate CT scanning appointments.
Data Source
AI summary
The present disclosure relates to a method for converting magnetic resonance imaging (MRI) to a computed tomography (CT) image using an artificial intelligence machine learning model, for use in ultrasound treatment device applications. The method includes acquiring training data including an MRI image and a CT image for machine learning; training an artificial neural network model using the training data, wherein artificial neural network model generates a CT image corresponding to the MRI image, and compares the generated CT image with the original CT image included in the training data; receiving an input MRI image to be converted to a CT image; splitting the input MRI image into a plurality of patches; generating patches of a CT image corresponding to the patches of the input MRI image using the trained artificial neural network model; and merging the patches of the CT image to generate an output CT image.